Accelerated Traffic Load Testing of Expansion Joints for ... · perform adequately under typical Bay Bridge traffic. The distresses observed on the Trelleborg unit under high loads

PREPARED FOR: California Department of Transportation Division of Research and Innovation Office of Roadway Research

PREPARED BY:

University of California Pavement Research Center

UC Davis, UC Berkeley

December 2011Research Report: UCPRC-RR-2011-06

AAAcccccceeellleeerrraaattteeeddd TTTrrraaaffffffiiiccc LLLoooaaaddd TTTeeessstttiiinnnggg ooofff EEExxxpppaaannnsssiiiooonnn JJJoooiiinnntttsss fffooorrr ttthhheee SSSeeelllfff---

AAAnnnccchhhooorrreeeddd SSSuuussspppeeennnsssiiiooonnn SSSeeeccctttiiiooonnn ooofff ttthhheee NNNeeewww SSSaaannn FFFrrraaannnccciiissscccooo–––OOOaaakkklllaaannnddd BBBaaayyy

BBBrrriiidddgggeee EEEaaasssttt SSSpppaaannn

Authors:D. Jones and R. Wu

Partnered Pavement Research Center (PPRC) Contract Strategic Plan Element 3.16: Bay Bridge Expansion Joint Testing Study

UCPRC-RR-2011-06 i

DOCUMENT RETRIEVAL PAGE Research Report: UCPRC-RR-2011-06

Title: Accelerated Traffic Load Testing of Expansion Joints for the Self-anchored Suspension Section of the New San Francisco–Oakland Bay Bridge East Span

Authors: David Jones and Rongzong Wu

Prepared for: Caltrans

FHWA No: CA122255A

Work submitted: 03-19-2012

DateDecember 2011

Strategic Plan Element No: 3.16

Status: Stage 6 Final

Version No.: 1

Abstract: A relatively unique opportunity was recently identified for accelerated traffic load testing of a new bridge expansion joint design. This study was part of the construction of the new East Span of the San Francisco–Oakland Bay Bridge and assessed whether these new Caltrans seismic expansion joints (which were designed to function in harmony with the bridge decks in the event of a high-magnitude earthquake) linking the Self-anchored Span with the Transition and Skyway spans would withstand truck traffic loading. A test structure incorporating one of the full-scale joints was constructed close to the actual bridge and tested with the California Department of Transportation / University of California Pavement Research Center Heavy Vehicle Simulator in a series of phases. A total of 1.36 million load repetitions, equating to about 46 million equivalent standard axle loads, were applied in seven phases during the three-month test. On completion of this testing, no structural damage was recorded by any of the Linear Variable Differential Transducers or strain gauges installed on the steel plates, steel frames, bolts, or washers. There was also no visible damage on any of these components. Excessive overloading with a 150 kN half-axle load in the last phase of the test caused some damage to the Trelleborg unit of the joint. This included abrasion, tearing, shoving and permanent deformation of the rubber inserts, and deformation and shearing of one of the steel supports directly under the wheel load. Based on the results of this limited testing, it was concluded that the Caltrans seismic expansion joint would perform adequately under typical Bay Bridge traffic. The distress observed to the Trelleborg unit under the high loads in the last phase of testing is unlikely to occur under normal traffic. However, the Trelleborg unit was found to be the weakest point of the expansion joint, as expected. On the actual bridge structure, these units will require periodic maintenance and replacement in line with manufacturer’s specifications. The findings from this study indicate that the Caltrans seismic expansion joint tested would be appropriate for typical Bay Bridge traffic. No seismic or structural testing was undertaken on the seismic expansion joint as part of this study and no recommendations toward its seismic or structural performance are made. Ride quality, skid resistance, and tire noise studies were carried out by Caltrans in a separate study and are reported on in separate Caltrans reports.

Keywords: San Francisco–Oakland Bay Bridge, Expansion Joint, Accelerated Traffic Load Test, Heavy Vehicle Simulator

Proposals for implementation: Caltrans seismic expansion joint as tested is expected to withstand normal traffic loading

Related documents: None

Signatures:

D. Jones 1st Author

J. Harvey Technical Review

D. Spinner Editor

J. Harvey Principal Investigator

T.J. Holland Caltrans Contract Manager

ii UCPRC-RR-2011-06

DISCLAIMER

The contents of this report reflect the views of the authors who are responsible for the facts and accuracy

of the data presented herein. The contents do not necessarily reflect the official views or policies of the

State of California or the Federal Highway Administration. This report does not constitute a standard,

specification, or regulation.

In this study, a new Caltrans seismic expansion joint was assessed for performance under accelerated

truck traffic loading only. No seismic or structural testing was undertaken on the seismic expansion joint

as part of this study and no recommendations toward its seismic or structural performance are made. Ride

quality, skid resistance, and tire noise studies were carried out by Caltrans in a separate study and are

reported on in separate Caltrans reports.

PROJECT OBJECTIVES

The objective of this study on accelerated traffic load testing of a Caltrans seismic expansion joint for the

self-anchored suspension section of the new San Francisco–Oakland Bay Bridge East Span Bay Bridge

was to provide a rapid indication of whether the expansion joint would perform adequately under typical

Bay Bridge vehicle loading. This was achieved through the following tasks:

1. Identify any fatal flaws in the design related to vehicle trafficking;

2. Determine how the expansion joint will fail under vehicle trafficking.

No seismic or structural testing was undertaken on the seismic expansion joint as part of this study and no

recommendations toward its seismic or structural performance are made. Ride quality, skid resistance,

and tire noise studies were carried out by Caltrans in a separate study and are reported on in separate

Caltrans reports.

UCPRC-RR-2011-06 iii

ACKNOWLEDGEMENTS

The University of California Pavement Research Center acknowledges the following individuals and

organizations who contributed to the project:

Dr. Joe Holland, Caltrans Division of Research and Innovation Mr. Jason Wilcox, Mr. Mike Whiteside, and Mr. Ric Maggenti, Caltrans Toll Bridge Program Mr. Hardik Patel and Mr. Charles Redfield, T.Y. Lin International Group The UCPRC Heavy Vehicle Simulator Crew under the leadership of Mr. Peter Millar

iv UCPRC-RR-2011-06

UCPRC-RR-2011-06 v

EXECUTIVE SUMMARY

A relatively unique opportunity was recently identified for accelerated traffic load testing of a new bridge

expansion joint design not previously used in California. This study was part of the construction of the

new East Span of the San Francisco–Oakland Bay Bridge and assessed whether the new expansion joints

(which were designed to function in harmony with the bridge decks in the event of a high-magnitude

earthquake) planned for linking the Self-anchored Span with the Transition and Skyway spans would

withstand truck traffic loading. A test structure incorporating one of the full-scale joints was constructed

close to the actual bridge and tested with the California Department of Transportation / University of

California Pavement Research Center Heavy Vehicle Simulator in a series of phases.

A total of 1.36 million load repetitions, equating to about 46 million equivalent standard axle loads on a

highway pavement, were applied in seven phases during the three-month test. On completion of this

testing, no structural damage was recorded by any of the Linear Variable Differential Transducers

(LVDTs) or strain gauges installed on the steel plates, steel frames, bolts, and washers. There was also no

visible damage on any of these components. Excessive overloading with a 150 kN half-axle load

(approximately four times the standard axle load) on an aircraft tire in the last phase of the test caused

some damage to the Trelleborg unit in the joint. The damage included abrasion, tearing, shoving and

permanent deformation of the rubber inserts, and deformation and shearing of one of the steel supports

directly under the wheel load.

Although no vehicle suspension dynamics (i.e., vehicle bounce) or speed effects were considered, based

on the results of this limited testing, it was concluded that the Caltrans seismic expansion joint would

perform adequately under typical Bay Bridge traffic. The distresses observed on the Trelleborg unit

under high loads in the last phase of testing are unlikely to occur under normal traffic. However, the

Trelleborg unit was found to be the weakest point of the expansion joint, as expected. On the actual

bridge structure, these units should be checked periodically to confirm the findings of this study, and to

assess any effects of higher speeds and vehicle dynamics that were not identified. The joints will require

periodic maintenance and replacement in line with manufacturer’s specifications.

The findings from this study indicate that the Caltrans seismic expansion joint tested would be

appropriate for typical Bay Bridge traffic.

No seismic or structural testing was undertaken and no recommendations toward the expansion joint’s

seismic or structural performance are made. Ride quality, skid resistance, and tire noise studies were

carried out by Caltrans in a separate study and are reported on in separate Caltrans reports.

vi UCPRC-RR-2011-06

UCPRC-RR-2011-06 vii

TABLE OF CONTENTS

EXECUTIVE SUMMARY ......................................................................................................................... v LIST OF TABLES ...................................................................................................................................... ix LIST OF FIGURES ..................................................................................................................................... x CONVERSION FACTORS .....................................................................................................................xiii 1. INTRODUCTION ............................................................................................................................. 1

1.1 Background ............................................................................................................................... 1 1.2 Project Objectives ...................................................................................................................... 2 1.3 Literature Review ...................................................................................................................... 3 1.4 Structure and Content of this Report ......................................................................................... 3 1.5 Measurement Units .................................................................................................................... 3

2. TEST STRUCTURE LOCATION, DESIGN, AND CONSTRUCTION ..................................... 5 2.1 Test Structure Location ............................................................................................................. 5 2.2 Test Structure Design ................................................................................................................ 6 2.3 Test Structure Construction ....................................................................................................... 6

3. HVS TEST PLAN, INSTRUMENTATION AND TEST CRITERIA .......................................... 9 3.1 HVS Test Protocols ................................................................................................................... 9 3.2 Test Plan .................................................................................................................................... 9 3.3 Instrumentation Plan and Test Section Layout ........................................................................ 10 3.4 Visual Assessments ................................................................................................................. 15 3.5 HVS Test Criteria .................................................................................................................... 15

3.5.1 Test Section Failure Criteria ....................................................................................... 15 3.5.2 Environmental Conditions .......................................................................................... 15 3.5.3 Loading Program ......................................................................................................... 15

4. HVS TEST DATA ........................................................................................................................... 17 4.1 Introduction ............................................................................................................................. 17 4.2 Phase 1.1: Fatal Flaw Assessment .......................................................................................... 17

4.2.1 Introduction ................................................................................................................. 17 4.2.2 Temperature ................................................................................................................ 17 4.2.3 Vertical Deflection ...................................................................................................... 18 4.2.4 Longitudinal Strain ..................................................................................................... 21 4.2.5 Visual Damage ............................................................................................................ 23 4.2.6 Phase Summary ........................................................................................................... 23

4.3 Phase 1.2: Load Response on the Center of the Steel Plate ................................................... 24 4.3.1 Introduction ................................................................................................................. 24 4.3.2 Temperature ................................................................................................................ 24 4.3.3 Vertical Deflection ...................................................................................................... 25 4.3.4 Longitudinal Strain ..................................................................................................... 27 4.3.5 Visual Damage ............................................................................................................ 28 4.3.6 Phase Summary ........................................................................................................... 30

4.4 Phase 1.3: Load Response Comparison at Center and Edge of the Steel Plate ...................... 30 4.4.1 Introduction ................................................................................................................. 30 4.4.2 Temperature ................................................................................................................ 31 4.4.3 Vertical Deflection ...................................................................................................... 32 4.4.4 Longitudinal Strain ..................................................................................................... 34 4.4.5 Visual Damage ............................................................................................................ 36 4.4.6 Phase Summary ........................................................................................................... 36

4.5 Phase 2.1: Edge Loading Test ................................................................................................ 37 4.5.1 Introduction ................................................................................................................. 37 4.5.2 Temperature ................................................................................................................ 37 4.5.3 Vertical Deflection ...................................................................................................... 38 4.5.4 Longitudinal Strain ..................................................................................................... 42

viii UCPRC-RR-2011-06

4.5.5 Visual Damage ............................................................................................................ 42 4.5.6 Phase Summary ........................................................................................................... 42

4.6 Phase 3.1: Edge Test with Impact Load and Unidirectional Traffic ...................................... 45 4.6.1 Introduction ................................................................................................................. 45 4.6.2 Temperature ................................................................................................................ 46 4.6.3 Vertical Deflection ...................................................................................................... 47 4.6.4 Longitudinal Strain ..................................................................................................... 48 4.6.5 Visual Damage ............................................................................................................ 49 4.6.6 Phase Summary ........................................................................................................... 49

4.7 Phase 3.2: Load Response with Impact Load ......................................................................... 49 4.7.1 Introduction ................................................................................................................. 49 4.7.2 Temperature ................................................................................................................ 50 4.7.3 Vertical Deflection ...................................................................................................... 50 4.7.4 Longitudinal Strain ..................................................................................................... 53 4.7.5 Visual Damage ............................................................................................................ 56 4.7.6 Phase Summary ........................................................................................................... 56

4.8 Phase 3.3: Edge Test with High Load .................................................................................... 57 4.8.1 Introduction ................................................................................................................. 57 4.8.2 Temperature ................................................................................................................ 57 4.8.3 Vertical Deflection ...................................................................................................... 58 4.8.4 Longitudinal Strain ..................................................................................................... 59 4.8.5 Visual Damage ............................................................................................................ 60 4.8.6 Phase Summary ........................................................................................................... 60

4.9 Static Responses for All Phases .............................................................................................. 62 4.9.1 Vertical Deflections .................................................................................................... 62 4.9.2 Longitudinal Strain ..................................................................................................... 63

4.10 Permanent Deformation on Trelleborg Unit ............................................................................ 65 5. CONCLUSIONS .............................................................................................................................. 71 6. REFERENCES ................................................................................................................................ 73 APPENDIX A: TEST STRUCTURE DESIGN AND INSTRUMENTATION ................................... 75

UCPRC-RR-2011-06 ix

LIST OF TABLES

Table 3.1: HVS Test Plan Summary ............................................................................................................ 9 Table 3.2: List of Instrumentation .............................................................................................................. 11 Table 3.3: Summary of HVS Loading Program ........................................................................................ 16 Table 4.1: Phase 1.1: Temperature Summary ........................................................................................... 18 Table 4.2: Phase 1.2: Temperature Summary ........................................................................................... 24 Table 4.3: Phase 1.3: Temperature Summary ........................................................................................... 31 Table 4.4: Average Peak Deflections for Different Lateral Wheel Positions on the Steel Plate ............... 32 Table 4.5: Average Peak Strains for Different Lateral Wheel Positions on the Steel Plate ....................... 35 Table 4.6: Phase 2.1: Temperature Summary ........................................................................................... 38 Table 4.7: Phase 3.1: Temperature Summary ........................................................................................... 46 Table 4.8: Phase 3.3: Temperature Summary ........................................................................................... 57

x UCPRC-RR-2011-06

LIST OF FIGURES

Figure 1.1: Schematic of the new East Span of the San Francisco–Oakland Bay Bridge. ........................... 2 Figure 2.1: Location of HVS test site (regional perspective). ...................................................................... 5 Figure 2.2: Location of HVS test site (local perspective). ........................................................................... 5 Figure 2.3: Location of HVS test site (site perspective). .............................................................................. 6 Figure 2.4: Initial excavation. ....................................................................................................................... 7 Figure 2.5: Formwork for test structure. ...................................................................................................... 7 Figure 2.6: Trelleborg installation. ............................................................................................................... 7 Figure 2.7: Steel plate installation. ............................................................................................................... 7 Figure 2.8: Completed steel plate installation. ............................................................................................. 7 Figure 2.9: Completed structure. .................................................................................................................. 7 Figure 2.10: Concrete pour problem on channel assembly structure. .......................................................... 8 Figure 2.11: Gap between steel plate and channel assembly (note concrete repair). ................................... 8 Figure 2.12: HVS on test structure prior to start of testing. ......................................................................... 8 Figure 3.1: Layout of instrumentation for testing on the center of the expansion joint. ............................ 12 Figure 3.2: Layout of instrumentation for testing on the edge of the expansion joint. .............................. 12 Figure 3.3: Relative location of HVS wheels for phases with channelized traffic. .................................... 13 Figure 3.4: Laser profilometer recording surface profile of the Trelleborg unit. ....................................... 13 Figure 3.5: General view of instruments on top of structure. ..................................................................... 13 Figure 3.6: LVDTs on channel assembly bolts. ......................................................................................... 13 Figure 3.7: LVDTs on edge of steel plate. ................................................................................................. 13 Figure 3.8: LVDTs on edge of steel plate. ................................................................................................. 14 Figure 3.9: LVDTs on Trelleborg unit. ...................................................................................................... 14 Figure 3.10: General view of instruments underneath steel plate. ............................................................. 14 Figure 3.11: LVDT on bottom of steel plate. ............................................................................................. 14 Figure 3.12: LVDT, strain gauge, and thermocouple on midpoint under steel plate. ................................ 14 Figure 3.13: LVDTs on channel assembly bolt washers. ........................................................................... 14 Figure 3.14: Dual truck tire configuration (note load calibration pad). ..................................................... 16 Figure 3.15: Aircraft tire (Boeing 737) configuration. ............................................................................... 16 Figure 4.1: Phase 1.1: Daily average temperatures and HVS testing schedule. ......................................... 18 Figure 4.2: Phase 1.1: Influence lines of vertical deflection for LVDTs on bolts. ..................................... 19 Figure 4.3: Phase 1.1: Influence lines of vertical deflection for LVDTs on bolts and washers. ................ 19 Figure 4.4: Phase 1.1: Influence lines of vertical deflection for LVDTs on steel plate. ............................ 20 Figure 4.5: Phase 1.1: History of peak deflections on bolts. ...................................................................... 20 Figure 4.6: Phase 1.1: History of peak deflections on bolts and washers. ................................................. 21 Figure 4.7: Phase 1.1: History of peak deflections on steel plate. .............................................................. 21 Figure 4.8: Phase 1.1: Influence lines of longitudinal strain at bottom of steel plate. ............................... 22 Figure 4.9: Phase 1.1: History of peak longitudinal strains at bottom of steel plate. ................................. 22 Figure 4.10: Phase 1.1: Rubber abrasion on Trelleborg unit after 100,000 load repetitions. .................... 23 Figure 4.11: Phase 1.2: Daily average temperatures and HVS testing schedule. ...................................... 25 Figure 4.12: Phase 1.2: History of peak deflections on bolts. .................................................................... 26 Figure 4.13: Phase 1.2: History of peak deflections on bolts and washers. ............................................... 26 Figure 4.14: Phase 1.2: History of peak deflections on steel plate. ............................................................ 27 Figure 4.15: Phase 1.2: Relationship between peak deflection and wheel load. ........................................ 27 Figure 4.16: Phase 1.2: History of longitudinal strains at bottom of steel plate......................................... 28 Figure 4.17: Phase 1.2: Relationship between peak strain and wheel load for SG #10. ............................ 29 Figure 4.18: Phase 1.2: Relationship between peak strain and wheel load for SG #11. ............................ 29 Figure 4.19: Phase 1.2: Relationship between peak strain and wheel load for SG #12. ............................ 30 Figure 4.20: Phase 1.3: Daily average temperatures and HVS testing schedule. ....................................... 31 Figure 4.21: Phase 1.3: Lowest peak deflections recorded on washers during traffic wander. .................. 33

UCPRC-RR-2011-06 xi

Figure 4.22: Phase 1.3: Highest peak deflections recorded on washers during traffic wander. ................. 33 Figure 4.23: Phase 1.3: Lowest peak deflections on steel plate during traffic wander. ............................. 34 Figure 4.24: Phase 1.3: Highest peak deflections on steel plate during traffic wander. ............................. 34 Figure 4.25: Phase 1.3: Lowest peak longitudinal strains on steel plate during traffic wander. ................ 35 Figure 4.26: Phase 1.3: Highest peak longitudinal strains on steel plate during traffic wander. ............... 36 Figure 4.27: Phase 1.3: Rubber particle accumulation on Trelleborg unit after 740,000 repetitions. ....... 36 Figure 4.28: Phase 2.1: Daily average temperatures and HVS testing schedule. ....................................... 38 Figure 4.29: Phase 2.1: Influence lines of vertical deflection for LVDTs on bolts. ................................... 39 Figure 4.30: Phase 2.1: Influence lines of vertical deflection for LVDTs on bolts and washers. .............. 39 Figure 4.31: Phase 2.1: Influence lines of vertical deflection for LVDTs on steel plate. .......................... 40 Figure 4.32: Phase 2.1: History of peak deflections on bolts. .................................................................... 40 Figure 4.33: Phase 2.1: History of peak deflections on bolts and washers. ............................................... 41 Figure 4.34: Phase 2.1: History of peak deflections at bottom of steel plate. ............................................ 41 Figure 4.35: Phase 2.1: Relationship between peak deflection and wheel load. ........................................ 42 Figure 4.36: Phase 2.1: History of longitudinal strains at bottom of steel plate......................................... 43 Figure 4.37: Phase 2.1: Relationship between peak strains and wheel load for SG #10. ........................... 43 Figure 4.38: Phase 2.1: Relationship between peak strains and wheel load for SG #11. ........................... 44 Figure 4.39: Phase 2.1: Relationship between peak strains and wheel load for SG #12. ........................... 44 Figure 4.40: Phase 2.1: Rubber particle accumulation on Trelleborg unit after 928,000 repetitions. ........ 45 Figure 4.41: Phase 3.1: Impact load from neoprene step. .......................................................................... 46 Figure 4.42: Phase 3.1: Impact load from wooden step. ............................................................................ 46 Figure 4.43: Phase 3.1: Daily average temperatures and HVS testing schedule. ....................................... 47 Figure 4.44: Phase 3.1: History of peak deflections on bolts. .................................................................... 47 Figure 4.45: Phase 3.1: History of peak deflections on bolts and washers. ............................................... 48 Figure 4.46: Phase 3.1: History of peak deflections at bottom of steel plate. ............................................ 48 Figure 4.47: Phase 3.1: History of peak longitudinal strains at bottom of steel plate. ............................... 49 Figure 4.48: Phase 3.2: Influence lines of vertical deflection for LVDTs on bolts. ................................... 50 Figure 4.49: Phase 3.2: Influence lines of vertical deflection for LVDTs on bolts and washers. .............. 51 Figure 4.50: Phase 3.2: Influence lines of vertical deflection for LVDTs on steel plate. .......................... 51 Figure 4.51: Phase 3.2: History of peak deflections on bolts. .................................................................... 52 Figure 4.52: Phase 3.2: History of peak deflections on bolts and washers. ............................................... 52 Figure 4.53: Phase 3.2: History of peak deflections at bottom of steel plate. ............................................ 53 Figure 4.54: Phase 3.2: Relationship between peak deflection and wheel load. ........................................ 53 Figure 4.55: Phase 3.2: Influence lines for longitudinal strains at bottom of steel plate. ........................... 54 Figure 4.56: Phase 3.2: History of peak longitudinal strains at bottom of steel plate. ............................... 54 Figure 4.57: Phase 3.2: Relationship between peak strains and wheel load for SG #10. ........................... 55 Figure 4.58: Phase 3.2: Relationship between peak strains and wheel load for SG #11. ........................... 55 Figure 4.59: Phase 3.2: Relationship between peak strains and wheel load for SG #12. ........................... 56 Figure 4.60: Phase 3.2: Rubber accumulation on Trelleborg unit after 1,191,000 repetitions. .................. 56 Figure 4.61: Phase 3.3: Daily average temperatures and HVS testing schedule. ....................................... 58 Figure 4.62: Phase 3.3: History of peak deflections on bolts. .................................................................... 58 Figure 4.63: Phase 3.3: History of peak deflections on bolts and washers. ............................................... 59 Figure 4.64: Phase 3.3: History of peak deflections at bottom of steel plate. ............................................ 59 Figure 4.65: Phase 3.3: History of peak longitudinal strains at bottom of steel plate. ............................... 60 Figure 4.66: Phase 3.3: Damage to steel rib of Trelleborg unit. ................................................................. 61 Figure 4.67: Phase 3.3: Deformation and shoving of rubber on Trelleborg unit. ....................................... 61 Figure 4.68: Phase 3.3: Tearing of rubber and accumulation of rubber particles in Trelleborg unit. ........ 61 Figure 4.69: Phase 3.3: Structure and Trelleborg unit after completion of testing. ................................... 61 Figure 4.70: Example daily variation for vertical deflections during Phase 1.1. ....................................... 62 Figure 4.71: History of daily maximum static vertical deflections on steel plate. ..................................... 63 Figure 4.72: Example daily variation for longitudinal strain at SG #10 during Phase 1.1. ........................ 63 Figure 4.73: Example daily variation for longitudinal strain at SG #11 during Phase 1.1. ........................ 64 Figure 4.74: Example daily variation for longitudinal strain at SG #12 during Phase 1.1. ........................ 64

xii UCPRC-RR-2011-06

Figure 4.75: History of daily maximum static longitudinal strains. ........................................................... 65 Figure 4.76: Maximum downward permanent deformation of the Trelleborg unit. .................................. 66 Figure 4.77: Average maximum downward permanent deformation of Trelleborg unit. .......................... 66 Figure 4.78: Phase 1.1: Contour plot of deformation (dual wheel, channelized on center). ...................... 67 Figure 4.79: Phase 1.2: Contour plot of deformation (dual wheel, channelized on center). ...................... 67 Figure 4.80: Phase 1.3: Contour plot of deformation (dual wheel, wander). ............................................. 68 Figure 4.81: Phase 2.1: Contour plot of deformation (dual wheel, channelized on edge). ........................ 68 Figure 4.82: Phase 3.1: Contour plot of deformation (dual wheel, channelized on edge). ........................ 69 Figure 4.83: Phase 3.2: Contour plot of deformation (dual wheel, channelized on edge with impact). .... 69 Figure 4.84: Phase 3.3 (20,000 reps): Contour plot of deformation (aircraft, channelized on edge). ........ 70 Figure 4.85: Phase 3.3 (final): Contour plot of deformation (aircraft, channelized on edge). ................... 70

UCPRC-RR-2011-06 xiii

CONVERSION FACTORS

SI* (MODERN METRIC) CONVERSION FACTORS

Symbol Convert From Convert To Symbol Conversion

LENGTH

mm millimeters inches in mm x 0.039

m meters feet ft m x 3.28

km kilometers mile mile km x 1.609

AREA

mm2 square millimeters square inches in2 mm2 x 0.0016

m2 square meters square feet ft2 m2 x 10.764

VOLUME

m3 cubic meters cubic feet ft3 m3 x 35.314

kg/m3 kilograms/cubic meter pounds/cubic feet lb/ft3 kg/m3 x 0.062

L liters gallons gal L x 0.264

L/m2 liters/square meter gallons/square yard gal/yd2 L/m2 x 0.221

MASS

kg kilograms pounds lb kg x 2.202

TEMPERATURE (exact degrees)

C Celsius Fahrenheit F °C x 1.8 + 32

FORCE and PRESSURE or STRESS

N newtons poundforce lbf N x 0.225

kPa kilopascals poundforce/square inch lbf/in2 kPa x 0.145

*SI is the symbol for the International System of Units. Appropriate rounding should be made to comply with Section 4 of ASTM E380.

(Revised March 2003)

UCPRC-RR-2011-06 1

1. INTRODUCTION

1.1 Background

The 13.5 km (8.4 mi.) San Francisco–Oakland Bay Bridge connects the city of San Francisco with the

East Bay cities of Oakland, Emeryville and Berkeley and is the start point of the Interstate 80 (I-80)

corridor. Based on data collected at the toll plaza, the bridge carries approximately 270,000 vehicles per

day (compared to the 100,000 carried by the Golden Gate Bridge), of which about three percent is truck

traffic. It currently consists of two separate bridges linked by a short tunnel on Yerba Buena Island. The

existing East Span, a steel box girder design constructed in 1936, was damaged by the 7.1-magnitude

Loma Prieta earthquake in 1989, during which a section of the top span, carrying the five westbound

lanes, collapsed onto the lower eastbound lanes. Although repairs were made and the bridge reopened

approximately one month after the earthquake, a complete seismic retrofit of the East Span to withstand

future similar or more severe earthquakes was not considered viable and construction of a new bridge was

approved. The West Span, which consists of two suspension bridge spans connected at a center

anchorage, was easier to retrofit to accommodate higher magnitude earthquakes. Retrofit work on this

part of the bridge was completed in 2004 and retrofit work on the West Approach was completed in 2009.

The new East Span consists of four separate parts (Figure 1.1):

The Oakland Touchdown, linking the new bridge to the existing I-80 infrastructure.

The Skyway, two side-by-side 1.9 km (1.2 mi.) long concrete spans (completed in 2008).

The Self-anchored Suspension Span (SAS), two side-by-side 470 m (1,540 ft.) long spans supported by a single tower, which is still under construction. It will be the longest bridge of its kind in the world. The span’s single 160 m (525 ft.) tall tower will match the height of the West Span’s towers. Its placement closer to the west end of the structure creates a distinctive asymmetrical design, with the single 1.6 km (1.0 mi.) long main cable presenting a sharper angle on the west side and a more sloping appearance on the east.

The Yerba Buena Island Transition Structure (YBITS), still under construction, will connect the Self-anchored Suspension Span to Yerba Buena Island (YBI), and will transition the new East Span’s side-by-side road decks to the upper and lower decks of the Yerba Buena Island tunnel and West Span.

The three radically different structures also required a new expansion joint design to link the three main

parts (Skyway, Self-anchored Suspension Span, and Yerba Buena Island Transition Structure) while

integrating with the seismic functioning of the entire bridge system. This new expansion joint was

subsequently designed by the California Department of Transportation (Caltrans) and T.Y. Lin

International Group and incorporates a Trelleborg Transflex 2400 expansion joint, a steel connector plate,

2 UCPRC-RR-2011-06

and fastening systems. The main focus of the design was to ensure that the joint acted in harmony with

the three structures during seismic activity. A secondary focus was the requirement that each lane

incorporate separate joints, in order to facilitate maintenance without major disruption to traffic. During

review of the joint design, questions were raised with regard to how the joints would perform under traffic

loading, given the focus on their seismic and maintenance requirements. An accelerated loading test,

using the California Department of Transportation / University of California Pavement Research Center

Heavy Vehicle Simulator (HVS) was therefore undertaken to provide a quick indication of how the joint

would perform under truck traffic.

Figure 1.1: Schematic of the new East Span of the San Francisco–Oakland Bay Bridge.

(http://en.wikipedia.org/wiki/File:SFOBBEastSpan.svg)

1.2 Project Objectives

The research presented in this report is part of Partnered Pavement Research Center Strategic Plan

Element 3.16 (PPRC SPE 3.16), titled “Bay Bridge Expansion Joint Testing Study,” undertaken for the

California Department of Transportation (Caltrans) by the University of California Pavement Research

Center (UCPRC). The objective of this project was to provide a rapid indication of whether the expansion

joint would perform adequately under typical Bay Bridge vehicle loading. This was achieved through the

following tasks:

Identify any fatal flaws in the seismic expansion joint design related to vehicle trafficking

Determine how the seismic expansion joint will fail under vehicle trafficking

UCPRC-RR-2011-06 3

No seismic or structural testing was undertaken on the seismic expansion joint as part of this study, and no

recommendations toward its seismic or structural performance are made. Ride quality, skid resistance,

and tire noise studies were carried out by Caltrans in a separate study and are reported on in separate

Caltrans reports.

1.3 Literature Review

A review of the literature found no published reference to any similar studies where equipment used for

accelerated pavement testing was used to test bridge expansion joints.

1.4 Structure and Content of this Report

This report presents an overview of the work carried out to meet the objectives of the study, and is

organized as follows:

Chapter 2 summarizes the HVS test structure location, design, and construction.

Chapter 3 details the HVS test plan, test section layout, instrumentation plan, and HVS test criteria.

Chapter 4 provides a summary of the HVS test data collected.

Chapter 5 provides conclusions from the study.

1.5 Measurement Units

Although Caltrans has recently returned to the use of U.S. standard measurement units, metric units have

always been used by the UCPRC in the design and layout of HVS test tracks, and for laboratory and field

measurements and data storage, to facilitate comparisons of data between accelerated pavement testing

studies worldwide. In this report, metric and English units (provided in parentheses after the metric units)

are provided in general discussion. In keeping with convention, only metric units are used in HVS data

analyses and reporting. A conversion table is provided on Page xiii at the beginning of this report.

4 UCPRC-RR-2011-06

UCPRC-RR-2011-06 5

2. TEST STRUCTURE LOCATION, DESIGN, AND CONSTRUCTION

2.1 Test Structure Location

The HVS test site was located on a temporarily vacant area close to the bridge construction offices at the

Port of Oakland, California (Figure 2.1 through Figure 2.3).

Figure 2.1: Location of HVS test site (regional perspective).

Figure 2.2: Location of HVS test site (local perspective).

Test site

Test site Bay Bridge tollgate

Caltrans office

6 UCPRC-RR-2011-06

Figure 2.3: Location of HVS test site (site perspective).

2.2 Test Structure Design

The Caltrans seismic expansion joint was designed by T.Y. Lin International Group, who also assisted

Caltrans with the design of a structure to house the joint for the accelerated load testing discussed in this

report. A copy of the design is provided in Appendix A. The dimensions matched those on the actual

bridge. Reinforced concrete approach slabs for the HVS wheels were included in the design.

2.3 Test Structure Construction

The test structure was built by Caltrans. Construction was started in March 2011, and completed in

July 2011. Photographs of the construction are provided in Figure 2.4 through Figure 2.9. Problems were

encountered with the concrete pour around the channel assembly structure (Figure 2.10), but this was

repaired prior to installation of the instruments and the start of testing (Figure 2.11). However, this

problem resulted in a gap between the steel plate and the channel assembly on the west side of the

structure (Figure 2.11). This gap could not be repaired without dismantling and reconstruction, which

prevented testing on this edge of the steel plate, since loading would have led to unrealistic responses

being recorded. The completed project with the HVS in place is shown in Figure 2.12.

I-80

Westbound I-80/I-880 connector

Burma Road

HVS

UCPRC-RR-2011-06 7

Figure 2.4: Initial excavation.

Figure 2.5: Formwork for test structure.

Figure 2.6: Trelleborg installation.

Figure 2.7: Steel plate installation.

Figure 2.8: Completed steel plate installation.

Figure 2.9: Completed structure.

8 UCPRC-RR-2011-06

Figure 2.10: Concrete pour problem on channel assembly structure.

(Concrete was repaired prior to HVS testing.)

Figure 2.11: Gap between steel plate and channel assembly (note concrete repair).

Figure 2.12: HVS on test structure prior to start of testing.

Gap

UCPRC-RR-2011-06 9

3. HVS TEST PLAN, INSTRUMENTATION, AND TEST CRITERIA

3.1 HVS Test Protocols

Heavy Vehicle Simulator (HVS) test section layout, test setup, trafficking, and measurements followed

standard University of California Pavement Research Center (UCPRC) protocols (1).

3.2 Test Plan

Two main tasks were identified for this accelerated load study:

Identify any fatal flaws in the seismic expansion joint design related to vehicle trafficking

Determine how the seismic expansion joint will fail under vehicle trafficking

A review of the literature found no published reference to any similar studies and given a testing period

limitation of three months, best use of this time was taken into consideration in preparing a test plan to

meet the study objectives. A phased approach was followed, starting with normal truck loads in the center

of the joint to identify any fatal flaws (Task 1), followed by incremental changes in loading and wheel

position to determine how the joint was likely to fail (Task 2). A test plan summary is provided in

Table 3.1.

Table 3.1: HVS Test Plan Summary

Phase No.

Test Section Location

Duration (days)

Wheel Loads (kN)

Repetitions Applied

1.1 1.2 1.3 2.1 3.1 3.2 3.3

Center Center

Center + Edge Edge Edge Edge Edge

30 6 7

11 3

15 15

1 day at 25, then 29 days at 40 1 day each at 25, 40, 60, 80, and 100, then back to 40 1 day each at 40, 100, and 80, then 4 days at 60 2 days at 40, 1 day each at 60, 80 and 100, then 6 days at 80 60, with impact load* 5 days each at 60, 80, and 100, all with impact load 1 day at 100, then 14 days at 150

518,000 120,000 120,000 189,000 23,000 240,000 150,000

- - 3 No test days 0 Total - 90 - 1,360,000

Test Section Numbering Phase

No Test Section

No 1.1 1.2 1.3 2.1 3.1 3.2 3.3

640HC 640HC

640HC-A 640HC-B 640HC-C 640HC-D 640HC-E

* Impact load was applied by forcing the HVS wheel over a step in the wheelpath created by either a 13 mm (1/2 in.) neoprene pad or 19 mm (3/4 in.) hardwood board.

In the first phase (Phase 1.1), testing at standard wheel loads in a channelized trafficking mode for four

weeks (i.e., equivalent to an 80 kN [18,000 lb] axle load) was included to identify any potential major

10 UCPRC-RR-2011-06

flaws in the design. The following phases would then evaluate the joint response under wandering traffic,

increasing wheel load, and different wheelpath (specifically along the edge of the joint). Assuming that no

damage was caused in the first two phases, the final phase would investigate impact loads and very high

wheel loads and tire pressures with a view to identifying the weakest point of the design.

Load variations on a single day were included in the study to establish relationships between wheel load

and structural response, and to identify any nonlinearity that might lead to structural damage.

3.3 Instrumentation Plan and Test Section Layout

The expansion joint was comprehensively instrumented to monitor status and responses under HVS

trafficking. Parameters monitored included ambient and steel plate temperatures, vertical deflections at

various locations, and longitudinal strain at the bottom of the steel plate. The instruments used and their

location on the bridge deck expansion joint are listed in Table 3.2. Layouts of the instrumentation for

testing on the center and edge of the joint are shown in Figure 3.1 and Figure 3.2, respectively.

Instruments #1 through #9 and Instrument #13 are Linear Variable Differential Transducers (LVDTs),

Instruments #10 through #12 are strain gauges, and Instruments #14 through #18 are thermocouples.

Location of LVDTs and strain gauges are also on the design drawings in Appendix A.

Standard HVS test sections were used for all testing. These are 8.0 m by 0.6 m (26.3 ft. by 2.0 ft.) for

channelized loading and 8.0 m by 1.0 m (26.3 ft. by 3.3 ft.) for loading with wheel wander.

Permanent deformation of the Trelleborg unit was measured with a laser profilometer. Figure 3.3 shows

the relative HVS wheel positions and location of profilometer measurements for the different testing

configurations. Stations 1 through 6 were inside the HVS wheelpath at some point during each phase

while Station 7 and Station 8 were outside the wheelpath at all times for all phases. Station 8 was

approximately 300 mm (12 in.) from the edge of the wheelpath and consequently no permanent

deformation should have been measured at this location. All surface profiles were measured in a

longitudinal direction (i.e., the trafficking direction) at 200 mm intervals in the transverse direction

(Figure 3.4). Daily change in surface elevation of the Trelleborg ribs was calculated by subtracting the

initial surface elevation from the deformed surface elevation.

Data from all instruments except the profilometer were collected continuously throughout the test.

Profilometer measurements were taken once a day while the HVS was stopped.

Photographs of the various instruments are shown in Figure 3.5 through Figure 3.13.

UCPRC-RR-2011-06 11

Table 3.2: List of Instrumentation

Instrument Number

Type Label Quantity Measured

Location

1 LVDT LVDT#1 Vertical Deflection On head of Bolt A5 for Phases 1.1 and 1.2, removed for Phase 1.3. On head of Bolt A3 for Phases 2 and 3.

2 LVDT LVDT#2 Vertical Deflection On head of Bolt B5 for Phases 1.1 and 1.2, removed for Phase 1.3. On head of Bolt B3 for Phases 2 and 3.

3 LVDT LVDT#3 Vertical Deflection Bottom washer under Bolt A4

4 LVDT LVDT#4 Vertical Deflection Bottom washer under Bolt B4

5 LVDT LVDT#5 Vertical Deflection Top of steel plate at outside edge

6 LVDT LVDT#6 Vertical Deflection Top of steel plate at inside edge for Phases 1.1 and 1.2, removed for Phase 1.3. Bottom of steel plate in the same horizontal location for Phases 2 and 3.

7 LVDT LVDT#7 Vertical Deflection Bottom of steel plate at midwidth

8 LVDT LVDT#8 Vertical Deflection On head of bolt at the connection between steel plate and Trelleborg; on head of Bolt 2 for Phases 1.1, 1.2, and 1.3; on head of Bolt 4 for Phases 2 and 3.

9 LVDT LVDT#9 Vertical Deflection On head of bolt at the connection between steel plate and Trelleborg; on head of Bolt 6 for Phases 1.1 and 1.2; removed for remaining phases.

10 Strain Gauge SG#10 Longitudinal Strain Bottom of steel plate at outside edge

11 Strain Gauge SG#11 Longitudinal Strain Bottom of steel plate at midwidth

12 Strain Gauge SG#12 Longitudinal Strain Bottom of steel plate at inside edge

13 LVDT LVDT#13 Vertical Deflection Bottom of steel plate at midwidth and midspan of the tunnel

14 Thermocouple TC-SG#10 Temperature Bottom of steel plate next to SG#10



17 Thermocouple TC-SG#10-S Temperature Surface of steel plate on top of SG#10

18 Thermocouple TC-Ambient Temperature Ambient air temperature next to steel plate

12 UCPRC-RR-2011-06

Figure 3.1: Layout of instrumentation for testing on the center of the expansion joint.

Figure 3.2: Layout of instrumentation for testing on the edge of the expansion joint.

UCPRC-RR-2011-06 13

Figure 3.3: Relative location of HVS wheels for phases with channelized traffic.

Figure 3.4: Laser profilometer recording surface profile of the Trelleborg unit.

Figure 3.5: General view of instruments on top of structure.

Figure 3.6: LVDTs on channel assembly bolts.

(Instruments #1 and #2)

Figure 3.7: LVDTs on edge of steel plate.

(Instrument #5)

14 UCPRC-RR-2011-06

Figure 3.8: LVDTs on edge of steel plate.

(Instrument #6)

Figure 3.9: LVDTs on Trelleborg unit.


Figure 3.10: General view of instruments underneath steel plate.

Figure 3.11: LVDT on bottom of steel plate.

(Instrument #13)

Figure 3.12: LVDT, strain gauge, and thermocouple on midpoint under steel plate.

(Instruments #7, #11, and #15)

Figure 3.13: LVDTs on channel assembly bolt washers.


UCPRC-RR-2011-06 15

3.4 Visual Assessments

Visual assessments of the Trelleborg unit, steel plate, channel assembly, bolts, and instruments were

undertaken on an hourly basis. Bolts and washers in the channel assembly were marked prior to the start

of testing and were checked for rotation on a daily basis.

3.5 HVS Test Criteria

3.5.1 Test Section Failure Criteria

No failure criteria were set for this study. Instead, all instrument data and profile measurements were

reviewed on a daily basis throughout the study and any unexpected distress/deformation/deflection

discussed with the design consultant.

3.5.2 Environmental Conditions

All testing was carried out under ambient conditions. Temperatures are summarized in Chapter 4.

3.5.3 Loading Program

The HVS loading program for each section is summarized in Table 3.3. Wheel loads applied are half axle

(i.e., the load applied by a 40 kN [9,000 lb] half axle is the same as that applied by an 80 kN [18,000 lb]

full-axle). Equivalent Standard Axle Loads (ESALs) were determined using the following Caltrans

pavement design conversion (Equation 3.1):

ESALs = (full axle load/80 kN)4.2 (3.1)

Most trafficking was applied in a channelized, bidirectional mode using dual wheel truck tires (Goodyear

G159 - 11R22.5- steel belt radial inflated to 720 kPa [104 psi]) with these exceptions:

Phase 1.3, which assessed the effects of bidirectional traffic wander using the dual tires (Figure 3.14),

Phase 3.1, which assessed the effects of an impact load in a unidirectional mode, and

Phase 3.3, which assessed the effects of very high bidirectional loads using an aircraft tire (Boeing 737, Figure 3.15) inflated to 1,380 kPa (200 psi).

Load was checked with a portable weigh-in-motion pad at the beginning of each test (Figure 3.14) and

after each load change.

All testing was carried out at a wheel speed of 9.5 km/h (5.9 mph).

16 UCPRC-RR-2011-06

Table 3.3: Summary of HVS Loading Program

Phase No. Wheel Load Repetitions

Applied ESALs*

kN lbs

1.1 25 40

5,625 9,000

20,000 498,000

3,000 498,000

1.2 25 40 60 80 100 40

5,625 9,000 13,500 18,000 22,500 9,000

20,000 20,000 20,000 20,000 20,000 20,000

3,000 20,000 110,000 368,000 938,000 20,000

1.3 40 100 80 60

9,000 22,500 18,000 13,500

20,000 20,000 20,000 60,000

20,000 938,000 368,000 329,000

2.1 40 60 80 100 80

9,000 13,500 18,000 22,500 18,000

36,000 20,000 20,000 19,000 94,000

36,000 110,000 368,000 891,000 1,728,000

3.1 60 13,500 23,000 126,000 3.2 60

80 100

13,500 18,000 22,500

91,000 69,000 80,000

500,000 1,268,000 3,754,000

3.3 100 150

22,500 33,750

20,000 130,000

938,000 33,489,000

Total 1,360,000 46,821,000 * Equivalent Standard Axle Load using Caltrans pavement design formula

Figure 3.14: Dual truck tire configuration (note load calibration pad).

Figure 3.15: Aircraft tire (Boeing 737) configuration.

UCPRC-RR-2011-06 17

4. HVS TEST DATA

4.1 Introduction

This chapter summarizes the data collected during the different phases of accelerated pavement testing.

Each phase is covered separately and includes discussion on temperature (measured with thermocouples at

various locations on and next to the test structure), vertical deflection (measured with LVDTs),

longitudinal strain (measured with strain gauges), visual observations, and a phase summary. Static

response and permanent deformation (measured with a laser profilometer) for all phases are discussed in

separate sections. Where appropriate, data plots are presented on the same scale for all phases to facilitate

comparisons.

4.2 Phase 1.1: Fatal Flaw Assessment

4.2.1 Introduction

The main tasks of this phase were identification of any major flaws in the seismic expansion joint design,

and evaluation of the strain and deflections caused by a small increase in wheel load. The test ran for 30

days. Test load on the first day was set at 25 kN (5,625 lbs) and thereafter at 40 kN (9,000 lbs). All

loading was applied to the center of the expansion joint in a bidirectional channelized mode.

4.2.2 Temperature

The average (daily, minimum, and maximum), lowest, and highest temperatures measured during

Phase 1.1 are summarized in Table 4.1. Daily average temperatures are plotted in Figure 4.1, with error

bars indicating minimum and maximum temperatures for the thermocouple located next to Strain

Gauge #11 (TC-SG#11). Average ambient temperatures were typical for the area and had a relatively

small diurnal range. Average daily maximum temperatures recorded on the steel plate were considerably

higher than the ambient temperatures (4°C to 6°C [7°F to 11°F]), but average daily minimum

temperatures were only slightly higher. This was attributed to heat absorption by the steel. There was

some difference between the temperatures recorded at the different strain gauges, with variation attributed

to partial shading or different/restricted air flow movements, especially for those thermocouples

underneath the structure. No extreme temperature events were recorded. It is unlikely that temperature

had any significant influence on the way that the bridge deck expansion joint components functioned

during this phase of testing.

18 UCPRC-RR-2011-06

Table 4.1: Phase 1.1: Temperature Summary

Thermocouple

Temperature (°C) Average of

Daily Average

Average of Daily

Minimum

Average of Daily

Maximum

Lowest Highest

Ambient TC-SG#10 TC-SG#10-S TC-SG#11 TC-SG#12

18 21 20 21 21

15 17 17 17 17

22 27 27 26 28

15 16 16 16 16

26 33 33 31 33

Thermocouple Temperature (°F)Ambient TC-SG#10 TC-SG#10-S TC-SG#11 TC-SG#12

64 69 69 69 71

59 63 62 63 63

72 81 81 79 83

58 61 60 61 61

79 91 92 87 91

0

5

10

15

20

25

30

35

40

8/5/11 8/10/11 8/15/11 8/20/11 8/25/11 8/30/11 9/4/11 9/9/11

Date

Tem

per

atu

re (

ºC)

0

50

100

150

200

250

300

350

400

450

500

550

600

Lo

ad R

epet

itio

ns

(x 1

,000

)

TC-AmbientTC-SG#10TC-SG#11TC-SG#12TC-SG#10-SNumber of load repetitions

40kN

25kN

Figure 4.1: Phase 1.1: Daily average temperatures and HVS testing schedule.

4.2.3 Vertical Deflection

Influence lines (or deflection bowls) from a single pass of the 40 kN wheel load for the LVDTs on the

bolts, washers, and steel plate are shown in Figure 4.2 through Figure 4.4, respectively. Vertical

deflections on the bolts and washers were very small (between zero and 0.05 mm) with deflection

increasing with proximity to the load wheels, as expected. Deflections measured on the washers were

slightly higher than those measured on the bolts. Deflections on the steel plate were higher than those on

the bolts and washers and ranged between 0.6 mm and 0.9 mm depending on location, with highest

deflections on the midpoints of edges of the steel plate and the midpoint of the steel plate.

UCPRC-RR-2011-06 19

Plots of the peak deflections measured on bolts, washers, and the steel plate for the duration of the phase

are shown in Figure 4.5 through Figure 4.7. Deflections increased with the change in wheel load as

expected. Thereafter, deflections recorded by each of the LVDTs remained constant, with no evidence of

damage accumulation with increasing load repetitions. Deflection did not appear to be influenced by

temperature, with variation attributed to slight variation in the actual load applied, which was always well

within the acceptable range for the hydraulic loading system on the HVS.

-0.30

-0.25

-0.20

-0.15

-0.10

-0.05

0.00

0.05

0.10

0.15

0.20

0.25

0.30

0 1 2 3 4 5 6 7 8

Wheel Position (m)

Ver

tic

al D

efle

cti

on

(m

m)

LVDT#1, On Bolt A5 LVDT#2, On Bolt B5

LVDT#8, On Bolt 2 near Trelleborg LVDT#9, On Bolt 6 near Trelleborg

Negative deflection = Downward movement/tightening of bolts

Trelleborg Steel Plate ConcreteConcrete

Figure 4.2: Phase 1.1: Influence lines of vertical deflection for LVDTs on bolts.

(Repetition #500,000, wheel load at 40 kN)

-0.30

-0.25

-0.20

-0.15

-0.10

-0.05

0.00

0.05

0.10

0.15

0.20

0.25

0.30

0 1 2 3 4 5 6 7 8Wheel Position (m)

Ver

tica

l D

efle

ctio

n (

mm

)


LVDT#3, On washer below Bolt A4 LVDT#4, On washer below Bolt B4

Trelleborg Steel Plate

Negative deflection = Downward movement/tightening of bolts

ConcreteConcrete

Figure 4.3: Phase 1.1: Influence lines of vertical deflection for LVDTs on bolts and washers.


20 UCPRC-RR-2011-06

-3.00

-2.75

-2.50

-2.25

-2.00

-1.75

-1.50

-1.25

-1.00

-0.75

-0.50

-0.25

0.00

0.25

0.50

0 1 2 3 4 5 6 7 8

Wheel Position (m)

Ver

tica

l D

efl

ecti

on

(m

m)

LVDT#5, Outside edge at midspan LVDT#6, Inside edge at midspan

LVDT#7, Bottom of steel plate at midwidth LVDT#13, Btm. of steel plate at midspan of tunnel

Negative deflection = Downward movement


Figure 4.4: Phase 1.1: Influence lines of vertical deflection for LVDTs on steel plate.


0.0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1.0

0 50 100 150 200 250 300 350 400 450 500 550

Accumulated Load Repetitions (x 1,000)

Pea

k D

efl

ec

tio

n (

mm

)

0

5

10

15

20

25

30

35

40

Ste

el P

late

Tem

per

atu

re (

ºC)

LVDT#1: On Bolt A5 LVDT#2: On Bolt B5LVDT#8: On Bolt 2 near Trelleborg LVDT#9: On Bolt 6 near TrelleborgSteel plate temperature

20kN

40kN

Figure 4.5: Phase 1.1: History of peak deflections on bolts.

UCPRC-RR-2011-06 21

0.0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1.0

0 50 100 150 200 250 300 350 400 450 500 550


Pea

k D

efl

ec

tio

n (

mm

)

0

5

10

15

20

25

30

35

40

Ste

el P

late

Te

mp

era

ture

(ºC

)

LVDT#1: On Bolt A5 LVDT#2: On Bolt B5LVDT#3: On washer below Bolt A4 LVDT#4: On washer below Bolt B4Steel plate temperature

20kN

40kN

Figure 4.6: Phase 1.1: History of peak deflections on bolts and washers.

0.0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1.0

0 50 100 150 200 250 300 350 400 450 500 550


Pea

k D

efle

ctio

n (

mm

)

0

5

10

15

20

25

30

35

40

Ste

el P

late

Tem

pe

ratu

re (

ºC)

LVDT#5: Outside edge at midspanLVDT#6: Inside edge at midspanLVDT#7: Bottom of steel plate at midwidthLVDT#13: Btm. of steel plate, midspan of tunnelSteel plate temperature

20kN

40kN

Figure 4.7: Phase 1.1: History of peak deflections on steel plate.

4.2.4 Longitudinal Strain

Influence lines (or strain bowls) from a single pass of the 40 kN wheel load (repetition #500,000) for the

strain gauges at the midpoints of the inside and outside edge and midpoint of the steel plate are shown in

Figure 4.8. Strains were very similar and ranged between 40 and 60 microstrain, with highest strain

recorded at the midpoint of the steel plate.

22 UCPRC-RR-2011-06

A plot of the peak strains for the three strain gauges for the duration of the phase is shown in Figure 4.9.

Peak strain increased with the change in wheel load as expected. After the load change, peak strain

recorded by each of the gauges remained constant, with no evidence of damage accumulation with

increasing load repetitions. There was no correlation between temperature and elastic response, although

some very small daily variation (~3 to 5 ) between early morning and early afternoon was observed

in the plots.

-25

0

25

50

75

100

125

150

175

200

0 1 2 3 4 5 6 7 8

Wheel Position (m)

Lo

ng

itu

din

al S

trai

n (

mic

ros

trai

n)

SG#10, Outside edge

SG#11, Midwidth

SG#12, Inside edge

Positive strain = Tension


Figure 4.8: Phase 1.1: Influence lines of longitudinal strain at bottom of steel plate.


0

20

40

60

80

100

120

140

160

180

200

0 50 100 150 200 250 300 350 400 450 500 550


Pea

k S

trai

n (

mic

rost

rain

)

0

5

10

15

20

25

30

35

40

Ste

el P

late

Tem

per

atu

re (

°C)

SG#10, Outside edge SG#11, Midwidth

SG#12, Inside edge Steel plate temperature near SG#11

20kN

40kN

Figure 4.9: Phase 1.1: History of peak longitudinal strains at bottom of steel plate.

UCPRC-RR-2011-06 23

4.2.5 Visual Damage

No visual damage was observed on the concrete structure, steel plate, bolts, or washers. However, some

wear, in the form of tire abrasion on the rubber sections, was observed on the Trelleborg unit after about

100,000 load repetitions. Small rubber particles started to accumulate on the steel ribs and in the bolt

recesses (Figure 4.10). Apart from some slight deformation (< 2.0 mm) on the rubber in the wheelpaths,

no other damage was observed on the Trelleborg unit. No rotation of the bolts or washers was observed.

Figure 4.10: Phase 1.1: Rubber abrasion on Trelleborg unit after 100,000 load repetitions.

4.2.6 Phase Summary

No apparent damage was observed at the end of Phase 1.1. The permanent vertical settlement of the

structure after testing was 0.2 mm, which was considered minimal and unlikely to influence joint

performance. No permanent deformation in the steel plate occurred during this phase, based on the strain

data recorded. Deflections and longitudinal strains induced by the 80 kN standard axle load (40 kN half

axle) at midspan of the steel plate were approximately 0.9 mm and 60 microstrain, respectively, and

remained constant throughout the phase (i.e., deflections and strains did not increase with increasing load

repetitions). The vertical deflections at the bolts and washers were less than 0.1 mm, with washers

deflecting a little more than the bolts. There was no distinct correlation between temperature and elastic

response in the steel plate; however, very small changes in peak strain between the coldest and warmest

periods each day were observed on the data plots on most days. Minor fluctuations in strain and deflection

measurements were most likely caused by very small fluctuations in the actual load applied by the HVS.

No fatal flaws in the expansion joint design were identified.

Start of rubber accumulation

24 UCPRC-RR-2011-06

4.3 Phase 1.2: Load Response on the Center of the Steel Plate

4.3.1 Introduction

Phase 1.2 assessed load response on the center of the steel plate by evaluating changes in strain and

deflection induced by increases in wheel load. The test ran for six days, with a load increase each day for

the first five days. Loads applied were 25 kN, 40 kN, 60 kN, 80 kN, and 100 kN, respectively. On the

sixth day, the load was changed back to 40 kN to assess recovery after the very high loads. All loading

was applied to the center of the expansion joint in a bidirectional channelized mode.

4.3.2 Temperature




Gauge #11 (TC-SG#11). Average ambient temperatures were again typical for the area and had a

relatively small diurnal range. Average daily minimum and maximum temperatures recorded on the steel

plate were similar to the ambient temperatures, except for the thermocouple at Strain Gauge #12, which

indicated a higher average daily maximum than the other measurement points (4°C [7°F]). This was

attributed to different/restricted air flow movements around the thermocouple (positioned underneath the

structure, furthest away from the opening). No extreme temperature events were recorded. It is unlikely

that temperature had any significant influence on the way that the bridge deck expansion joint components

functioned during this phase of testing.


Thermocouple


Daily Average

Average of Daily

Minimum

Average of Daily

Maximum

Lowest Highest


18 19 19 20 21

15 17 16 17 17

22 23 23 23 27

14 15 15 16 16

27 26 27 26 35


64 66 66 67 69

59 62 62 63 63

71 74 74 74 81

56 59 58 60 60

81 80 81 79 95

UCPRC-RR-2011-06 25

0

5

10

15

20

25

30

35

40

9/4/11 9/6/11 9/8/11 9/10/11 9/12/11 9/14/11 9/16/11 9/18/11

Date

Tem

pe

ratu

re (

ºC)

0

25

50

75

100

125

150

Lo

ad R

epet

itio

ns

(x 1

,000

)


25kN 40kN 60kN 80kN 100kN 40kN



Plots of the peak deflections measured on bolts, washers, and the steel plate for the duration of Phase 1.2

are shown in Figure 4.12 through Figure 4.14. Deflections increased with the change in wheel load as

expected, but were still very small, with deflection on the bolts, washers, and steel plate ranging between

zero and 0.25 mm, zero and 0.15 mm, and 0.4 mm and 2.2 mm, respectively depending on load and sensor

location. The relationship between peak deflection and load was linear for loads between 20 kN and

80 kN, but showed marginal non-linearity for the 100 kN load (example for LVDT #5 in Figure 4.15).

The reason for this was not investigated given the very small difference and that the bridge deck

expansion joints would not be subjected to loads of this magnitude. After each load change, deflections

recorded by each of the LVDTs remained constant until the next load change. There was no evidence of

damage accumulation with increasing load repetitions. Deflections recorded during the 40 kN loading on

the second and sixth days were essentially the same (0.01 mm lower on the sixth day), indicating that no

permanent damage was caused by the very high wheel loads. Daily temperature change appeared to result

in very small daily variations in deflection (<0.01 mm on the bolts and washers and 0.1 mm on the steel

plate), especially at the lower loads.

26 UCPRC-RR-2011-06

0.0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1.0

0 10 20 30 40 50 60 70 80 90 100 110 120 130


Pea

k D

efle

ctio

n (

mm

)

0

5

10

15

20

25

30

35

40

Ste

el

Pla

te T

emp

era

ture

(ºC

)

LVDT#1: On Bolt A5 LVDT#2: On Bolt B5LVDT#8: On Bolt 2 near Trelleborg LVDT#9: On Bolt 6 near TrelleborgSteel plate temperature



0.0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1.0

0 10 20 30 40 50 60 70 80 90 100 110 120 130


Pea

k D

efl

ecti

on

(m

m)

0

5

10

15

20

25

30

35

40

Ste

el P

late

Tem

per

atu

re (

ºC)




UCPRC-RR-2011-06 27

0.0

0.5

1.0

1.5

2.0

2.5

3.0

3.5

0 10 20 30 40 50 60 70 80 90 100 110 120 130


Pea

k D

efle

ctio

n (

mm

)

0

5

10

15

20

25

30

35

40

Ste

el P

late

Te

mp

erat

ure

(ºC

)

LVDT#5: Outside edge at midspan LVDT#6: Inside edge at midspan

LVDT#7: Bottom of steel plate at midwidth LVDT#13: Btm. of steel plate, midspan of tunnel

Steel plate temperature


Figure 4.14: Phase 1.2: History of peak deflections on steel plate.

y = 0.0262x - 0.1223

R2 = 0.9932

0.0

0.5

1.0

1.5

2.0

2.5

3.0

0 10 20 30 40 50 60 70 80 90 100 110 120

HVS Wheel Load (kN)

Pea

k V

erti

cal D

efle

ctio

n (

mm

)

Response for loads between 25 and 80kNResponse at 100kN40kN after response testLinear Regression

LVDT #5

Figure 4.15: Phase 1.2: Relationship between peak deflection and wheel load.

(LVDT #5, midspan, outside edge of steel plate)



Peak strain increased with the change in wheel load as expected, with highest strains recorded at the

midpoint of the steel plate. Strains measured at the midpoint of the inside edge of the steel plate were

slightly higher than those measured at the midpoint of the outside edge. This was attributed to the

28 UCPRC-RR-2011-06

wheelpath being closer to the sensor on the inside edge. The difference in strain between the three sensors

increased with increasing wheel load, as expected, showing a general linear trend. After each load

change, peak strain recorded by each of the gauges remained constant, with no evidence of damage

accumulation with increasing load repetitions. The relationship between peak strain and load was linear

for all three strain gauges for loads between 20 kN and 80 kN, but again showed marginal non-linearity

for the 100 kN load (Figure 4.17 through Figure 4.19 for the three strain gauges). The reason for this was

not investigated given the very small difference and that the bridge deck expansion joints would not be

subjected to loads of this magnitude. Strains recorded during the 40 kN loading on the sixth day were

slightly lower (approximately 5 ) than those recorded on the second day, indicating that no permanent

damage was caused by the very high loads. Daily variation in peak strain followed daily temperature

change, similar to the earlier phases.

4.3.5 Visual Damage

No visual damage was observed on the concrete structure, steel plate, bolts, or washers at the end of this

phase. Tire abrasion wear on the Trelleborg unit, discussed in Section 4.2.5, continued with additional

accumulations of rubber particles. No further deformation or other damage was observed on the

Trelleborg unit. No rotation of the bolts or washers was observed.

0

20

40

60

80

100

120

140

160

180

200

0 10 20 30 40 50 60 70 80 90 100 110 120 130


Pea

k S

tra

in (

mic

rost

rain

)

0

5

10

15

20

25

30

35

40

Ste

el P

late

Te

mp

era

ture

(ºC

)

SG#10, Outside edge SG#11, Midwidth

SG#12, Inside edge Steel plate temperature near SG#11


Figure 4.16: Phase 1.2: History of longitudinal strains at bottom of steel plate.

UCPRC-RR-2011-06 29

SG #10

y = 1.3699x - 4.7517

R2 = 0.993

0

20

40

60

80

100

120

140

160

180

200

0 10 20 30 40 50 60 70 80 90 100 110 120

HVS Wheel Load (kN)

Pea

k L

on

git

ud

inal

Str

ain

(m

icro

stra

in)

Response for loads between 25kN and 80kN100kN40kN after response testLinear Regression

Figure 4.17: Phase 1.2: Relationship between peak strain and wheel load for SG #10.

(SG #10, outside edge of steel plate)

SG #11

y = 1.5382x - 1.3352

R2 = 0.9929

0

20

40

60

80

100

120

140

160

180

200

0 10 20 30 40 50 60 70 80 90 100 110 120

HVS Wheel Load (kN)

Pea

k L

on

git

ud

ina

l S

tra

in (

mic

ros

tra

in)

Response for loads between 25kN and 80kN

100kN

40kN after response test

Linear Regression


(SG #11, midspan, outside edge of steel plate)

30 UCPRC-RR-2011-06

SG #12

y = 1.1315x - 3.3091

R2 = 0.9924

0

20

40

60

80

100

120

140

160

180

200

0 10 20 30 40 50 60 70 80 90 100 110 120

HVS Wheel Load (kN)

Pea

k L

on

git

ud

inal

Str

ain

(m

icro

stra

in)


100kN


Linear Regression



4.3.6 Phase Summary

No damage was observed at the end of Phase 1.2. No permanent deformation in the steel plate occurred

during this phase, based on the strain data recorded. Increases in peak deflection and peak strain showed a

linear relationship with increasing load. The maximum deflection and maximum strain recorded was

2.3 mm and 135 respectively, both at the midpoint of the steel plate, under the 100 kN wheel load.

Changes in deflection and strain with increasing wheel load showed similar trends. Very small daily

variations in peak deflection and peak strain were consistent with daily temperature change on the data

plots. Minor fluctuations in strain and deflection measurements were again likely caused by very small

fluctuations in the actual load applied by the HVS.

4.4 Phase 1.3: Load Response Comparison at Center and Edge of the Steel Plate

4.4.1 Introduction

Phase 1.3 compared load response at the center and edge of the steel plate during traffic wander by

evaluating changes in strain and deflection induced by increases in wheel load. The test ran for seven

days, with a load change each day for the first three days (40 kN, 100 kN and 80 kN), followed by four

days at 60 kN. All loading was applied in a bidirectional mode using a stepwise normally distributed

wander pattern over a 1.0 m wide test track.

UCPRC-RR-2011-06 31

4.4.2 Temperature







again indicated a higher average daily maximum than the other measurement points. No extreme

temperature events were recorded. It is unlikely that temperature had any significant influence on the way

that the bridge deck expansion joint components functioned during this phase of testing.


Thermocouple


Daily Average

Average of Daily

Minimum

Average of Daily

Maximum

Lowest Highest


21 22 22 22 26

16 17 17 18 18

25 26 27 25 31

15 16 16 17 17

29 29 29 28 36


69 72 72 72 78

60 63 62 64 65

78 79 81 77 89

59 61 60 62 62

84 83 85 82 96

0

5

10

15

20

25

30

35

40

9/14/11 9/15/11 9/16/11 9/17/11 9/18/11 9/19/11 9/20/11 9/21/11 9/22/11 9/23/11

Date

Tem

per

atu

re (

ºC)

0

25

50

75

100

125

150

Lo

ad

Rep

etit

ion

s (

x 1

,00

0)


100kN40kN 80kN 60kN


32 UCPRC-RR-2011-06


Plots of the lowest (wheel wander point furthest from the sensor) and highest (wheel wander point closest

to the sensor) peak deflections measured on washers and the steel plate for the duration of Phase 1.3 are

shown in Figure 4.21 through Figure 4.24, respectively. Plots for deflection on the bolts are not shown

given the very small movements that were measured on them (<0.02 mm). Average deflections for the

sensors on the steel plate are also summarized in Table 4.4 together with a ratio between the highest and

lowest deflection recorded on each sensor. The difference in deflection on the washers during traffic

wander with a 40 kN wheel load was very small (~0.01 mm), with the washer closest to the outside edge

of the channel assembly having a slightly higher deflection than the one on the inside edge of the

assembly. It increased slightly with the 60 kN wheel load (~0.02 mm) and a little more for the 80 kN load

(~0.04 mm). At 100 kN, the difference between the lowest and highest deflections measured was a little

more noticeable at about 0.08 mm; however, the difference between the two washers for the same traffic

pass was about 0.2 mm at the higher load.

Deflections on the steel plate were a little higher, with bigger differences and larger variation between the

different sensor locations. The sensor furthest away from the wheel was most affected by wheel position.

At the 40 kN wheel load, differences in deflection ranged between 0.1 mm (LVDT #7 at the midpoint on

the bottom of the steel plate and LVDT #13 on the bottom of the steel plate in the midspan of the tunnel)

to 0.5 mm on the outside edge at the midspan (LVDT #5). The effect of increasing wheel load was similar

to that observed for the LVDTs on the washers, except that the deflections were higher. The effect of

wheel wander was most noticeable on the outside edge of the steel plate, with a variation of about 2.3 mm

(or 2.75 times higher) at the 100 kN wheel load, between the time that the wheel was closest to the sensor

and the time that it was furthest away. By comparison, variation in the middle of the steel plate during the

two extremes of wheel position was about 0.35 mm. This difference in deflection range was also partially

attributed to there being less support under the outside edge of the steel plate, compared to the midpoint

and inside edge, resulting from a construction problem (see Section 2.3).

Table 4.4: Average Peak Deflections for Different Lateral Wheel Positions on the Steel Plate

Test Load (kN)

Lowest Peak Deflection (mm)

Highest Peak Deflection (mm)

Ratio of Highest to Lowest

LVDT #5

LVDT #7

LVDT #13

LVDT #5

LVDT #7

LVDT #13

LVDT #5

LVDT #7

LVDT #13

40 60 80

100

0.31 0.45 0.64 0.83

0.63 0.91 1.25 1.57

0.51 0.74 1.00 1.27

0.80 1.25 1.79 2.29

0.72 1.10 1.52 1.94

0.58 0.89 1.25 1.59

2.58 2.81 2.80 2.75

1.14 1.21 1.22 1.23

1.14 1.20 1.24 1.25

The trends in deflection at different wheel loads showed similar linearity to that observed during

Phase 1.2. After each load change, deflections recorded by each of the LVDTs remained constant until

UCPRC-RR-2011-06 33

the next load change. There was no evidence of damage accumulation with increasing load repetitions

and the deflections recorded during wander, even at the high wheel load, were not considered to be

detrimental to the longer-term performance of the steel plate. Daily temperature change appeared to result

in very small daily variations in deflection (<0.01 mm), especially at the lower loads.

0.0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1.0

0 10 20 30 40 50 60 70 80 90 100 110 120 130


Lo

we

st P

eak

De

fle

cti

on

(m

m)

0

5

10

15

20

25

30

35

40

Ste

el P

late

Te

mp

era

ture

(ºC

)

LVDT#3: On washer below Bolt A4LVDT#4: On washer below Bolt B4Steel plate temperature

40kN 100kN 80kN 60kN

Figure 4.21: Phase 1.3: Lowest peak deflections recorded on washers during traffic wander.

0.0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1.0

0 10 20 30 40 50 60 70 80 90 100 110 120 130


Hig

he

st P

eak

De

fle

ctio

n (

mm

)

0

5

10

15

20

25

30

35

40

Ste

el

Pla

te T

emp

erat

ure

(ºC

)

LVDT#3: On washer below Bolt A4LVDT#4: On washer below Bolt B4Steel plate temperature

40kN 100kN 80kN 60kN

Figure 4.22: Phase 1.3: Highest peak deflections recorded on washers during traffic wander.

34 UCPRC-RR-2011-06

0.0

0.5

1.0

1.5

2.0

2.5

3.0

3.5

0 10 20 30 40 50 60 70 80 90 100 110 120 130


Lo

we

st

Pe

ak

De

flec

tio

n (

mm

)

0

5

10

15

20

25

30

35

40

Ste

el

Pla

te T

emp

erat

ure

(ºC

)

LVDT#5: Outside edge at midspanLVDT#7: Bottom of steel plate at midwidthLVDT#13: Btm. of steel plate, midspan of tunnelSteel plate temperature

40kN 100kN 80kN

60kN

Figure 4.23: Phase 1.3: Lowest peak deflections on steel plate during traffic wander.

0.0

0.5

1.0

1.5

2.0

2.5

3.0

3.5

0 10 20 30 40 50 60 70 80 90 100 110 120 130


Hig

hes

t P

eak

Def

lec

tio

n (

mm

)

0

5

10

15

20

25

30

35

40

Ste

el

Pla

te T

emp

era

ture

(ºC

)

LVDT#5: Outside edge at midspanLVDT#7: Bottom of steel plate at midwidthLVDT#13: Btm. of steel plate, midspan of tunnelSteel plate temperature

40kN 100kN 80kN

60kN

Figure 4.24: Phase 1.3: Highest peak deflections on steel plate during traffic wander.


Plots of the lowest (wheel wander point furthest from the sensor) and highest (wheel wander point closest

to the sensor) peak strains measured on the steel plate for the duration of Phase 1.3 are shown in

Figure 4.25 and Figure 4.26, respectively. Average strains are also summarized in Table 4.5 together with

a ratio between the highest and lowest peak strain recorded on each sensor at each wheel load. Changes in

strain at different wheel loads and different wheel positions were consistent with the changes observed in

UCPRC-RR-2011-06 35

deflection (discussed in Section 4.4.3). Differences in strains measured by the sensors on the edges of the

steel plate varied significantly more with changing wheel position than the sensor at the midpoint (ratio of

highest to lowest strain of between 2.1 and 2.5 for the outside edge compared to between 1.4 and 1.5 for

the midpoint).

Table 4.5: Average Peak Strains for Different Lateral Wheel Positions on the Steel Plate

Test Load (kN)

Lowest Peak Strain ()

Highest Peak Strain () Ratio of Highest to Lowest

SG#10 SG#11 SG#12 SG#10 SG#11 SG#12 SG#10 SG#11 SG#12 40 60 80

100

23 33 48 62

38 55 78 99

35 52 74 97

46 71 97

127

55 84 112 143

89 129 174 216

2.0 2.1 2.0 2.1

1.5 1.5 1.4 1.4

2.5 2.5 2.3 2.2

The trends in peak strain at different wheel loads showed similar linearity to that observed during

Phase 1.2. After each load change, peak strains recorded at each of the strain gauges remained constant

until the next load change. There was no evidence of damage accumulation with increasing load

repetitions and the peak strains recorded during wander, even at the high wheel load, were not considered

to be detrimental to the longer-term performance of the steel plate. Daily temperature cycles appeared to

result in very small daily variations in peak strain (~ 8 ), especially at the lower wheel loads.

0

50

100

150

200

250

300

350

0 10 20 30 40 50 60 70 80 90 100 110 120 130


Lo

wes

t P

eak

Str

ain

(m

icro

str

ain

)

0

5

10

15

20

25

30

35

40

Ste

el

Pla

te T

em

pe

ratu

re (

ºC)

SG#10: Outside edge SG#11: Midwidth

SG#12: Inside edge Steel plate temperature near SG#11

40kN

100kN 80kN 60kN

Figure 4.25: Phase 1.3: Lowest peak longitudinal strains on steel plate during traffic wander.

36 UCPRC-RR-2011-06

0

50

100

150

200

250

300

350

0 10 20 30 40 50 60 70 80 90 100 110 120 130


Hig

he

st P

eak

Str

ain

(m

icro

str

ain

)

0

5

10

15

20

25

30

35

40

Ste

el P

late

Tem

per

atu

re (

ºC)

SG#10: Outside edge SG#11: Midwidth

SG#12: Inside edge Steel plate temperature near SG#1140kN

100kN 80kN 60kN

Figure 4.26: Phase 1.3: Highest peak longitudinal strains on steel plate during traffic wander.

4.4.5 Visual Damage


phase. Tire abrasion wear on the Trelleborg unit continued with additional accumulations of rubber

particles (Figure 4.27). No further deformation or other damage was observed on the Trelleborg unit. No

rotation of the bolts or washers was observed.

Figure 4.27: Phase 1.3: Rubber particle accumulation on Trelleborg unit after 740,000 repetitions.

4.4.6 Phase Summary

No damage was observed at the end of Phase 1.3 and based on the deflection and strain data recorded, no

permanent deformation in the steel plate occurred. Peak strain and deflection at any time was influenced

by the position of the wheels in the wander pattern, as expected. Sensors furthest away from the wheels

Rubber accumulation

UCPRC-RR-2011-06 37

(i.e., on the edge of the steel plate) had larger differences between the lowest and highest deflection and

strain (ratio of ~2.5) compared to the sensors inside the wheelpath (i.e., at the midpoint of the steel plate),

which had highest strain to lowest strain ratios of about 1.5. Increases in peak deflection and peak strain

continued to show a linear relationship with increasing load. Very small daily variations in peak

deflection and peak strain were consistent with daily temperature change on the data plots. Minor

fluctuations in strain and deflection measurements were again likely caused by very small fluctuations in

the actual load applied by the HVS. Based on the results and observations in this phase, it was concluded

that there was no significant difference in the measurements recorded during traffic wander compared to

those recorded during channelized traffic, and that wander had very little effect on the behavior of the

expansion joint. Consequently all further testing was carried out in a channelized mode as this was

considered more likely to induce damage given the concentrated nature of the loading.

4.5 Phase 2.1: Edge Loading Test

4.5.1 Introduction

During Phase 2.1, testing was carried out on the edge of one of the expansion joints as shown in

Figure 3.3. The term “edge” is used because the wheelpath for this phase was closer to the edge of one of

the two steel plates, despite being closer to the center of the entire lane width. Normal trafficking on the

actual bridge would not occur in this way except when vehicles change lanes. The objective in this phase

was to assess whether trafficking at higher loads on the edge of the steel plate would cause the expansion

joint to behave differently compared to the central loading in Phases 1.1 and 1.2, and to determine if any

new damage was caused. The test ran for 11 days with a range of wheel loads (2 days at 40 kN, 1 day each

at 60 kN, 80 kN and 100 kN, then 6 days at 80 kN) to allow monitoring of changes in response. The

longer period of testing at the end of the phase was conducted at a higher wheel load than previous phases

to further explore likely modes of failure. All loading was applied in a bidirectional channelized mode.

4.5.2 Temperature







again indicated a higher average daily maximum than the other measurement points. No extreme



38 UCPRC-RR-2011-06


Thermocouple


Daily Average

Average of Daily

Minimum

Average of Daily

Maximum

Lowest Highest


19 20 20 21 23

16 17 17 18 19

25 24 25 24 30

15 16 15 16 17

33 29 30 29 38


66 68 68 69 73

61 63 62 64 66

76 76 77 75 85

59 60 60 61 62

91 85 87 84

100

0

5

10

15

20

25

30

35

40

9/22/11 9/24/11 9/26/11 9/28/11 9/30/11 10/2/11 10/4/11 10/6/11

Date

Tem

per

atu

re (

ºC)

0

25

50

75

100

125

150

175

200

225

250

Lo

ad R

epet

itio

ns

(x 1

,000

)


60kN40kN 80kN 100kN

80kN



Influence lines (or deflection bowls) from a single pass of the 80 kN wheel load for the LVDTs on the

bolts, washers, and steel plate are shown in Figure 4.29 through Figure 4.31, respectively. Vertical

deflections on the bolts and washers were higher than those recorded in a similar test in Phase 1.1 at

40 kN, as expected because of the higher load. Deflections showed similar trends to those observed in

Phase 1.1 and were still considered to be very small (between 0.02 mm and 0.12 mm) with deflection

increasing or decreasing with proximity to the wheel, as expected. Deflections measured on the washers

were similar to those measured on the bolts. Deflections on the steel plate were again significantly higher

than those on the bolts and washers, ranging between 0.75 mm and 2.0 mm depending on location of the

sensor, with highest deflections on the midpoint of the inside edge of the steel plate (closest sensor to the

wheel) and midpoint of the steel plate. The deflections on the bolts and washers on the channel assembly

UCPRC-RR-2011-06 39

and on the steel plate briefly changed from a negative deflection to a positive deflection and then back to a

zero deflection when the wheels moved from the steel plate to the concrete, indicating a small recovery

“bounce” after the load was removed. The movement was very small (total of 0.025 mm, 0.06 mm, and

0.08 mm on the bolts, washers, and steel plate respectively) and was not considered to be of any

consequence in terms of long-term performance.

-0.30

-0.25

-0.20

-0.15

-0.10

-0.05

0.00

0.05

0.10

0.15

0.20

0.25

0.30


Ver

tica

l D

efle

ctio

n (

mm

)

LVDT#1, On Bolt A3 LVDT#2, On Bolt B3 LVDT#8, On Bolt 4 near Trelleborg


Negative deflection = Downward movement/Tightening of bolts

ConcreteConcrete



-0.30

-0.25

-0.20

-0.15

-0.10

-0.05

0.00

0.05

0.10

0.15

0.20

0.25

0.30


Ver

tica

l D

efle

ctio

n (

mm

)


LVDT#3, On washer below Bolt A4 LVDT#4, On washer below Bolt B4



ConcreteConcrete



40 UCPRC-RR-2011-06

-3.00

-2.75

-2.50

-2.25

-2.00

-1.75

-1.50

-1.25

-1.00

-0.75

-0.50

-0.25

0.00

0.25

0.50

0 1 2 3 4 5 6 7 8

Wheel Position (m)

Ver

tica

l D

efle

ctio

n (

mm

)








are shown in Figure 4.32 through Figure 4.34, respectively. Deflections increased with the change in

wheel load as expected, and were consistent with observations from previous phases. The relationship

between peak deflection and load was linear for all loads (example for LVDT #5 in Figure 4.35). After the

load change, deflections recorded by each of the LVDTs remained constant, with no evidence of damage

accumulation with increasing load repetitions. Deflection did not appear to be influenced by temperature.

0.0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1.0

0 20 40 60 80 100 120 140 160 180 200


Pe

ak D

efl

ect

ion

(m

m)

0

5

10

15

20

25

30

35

40

Ste

el

Pla

te T

em

per

atu

re (

ºC)

LVDT#1: On Bolt A3LVDT#2: On Bolt B3LVDT#8: On Bolt 4 near TrelleborgSteel plate temperature

40kN 60kN 80kN 80kN100kN


UCPRC-RR-2011-06 41

0.0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1.0

0 20 40 60 80 100 120 140 160 180 200


Pea

k D

efle

cti

on

(m

m)

0

5

10

15

20

25

30

35

40

Ste

el P

late

Tem

pe

ratu

re (

ºC)

LVDT#1: On Bolt A3LVDT#2: On Bolt B3LVDT#3: On washer below Bolt A4LVDT#4: On washer below Bolt B4Steel plate temperature



0.0

0.5

1.0

1.5

2.0

2.5

3.0

3.5

0 20 40 60 80 100 120 140 160 180 200


Pea

k D

efle

ctio

n (

mm

)

0

5

10

15

20

25

30

35

40

Ste

el P

late

Tem

per

atu

re (

ºC)

LVDT#5: Outside edge at midspan LVDT#7: Bottom of steel plate at midwidth

LVDT#13: Btm. of steel plate, midspan of tunnel Steel plate temperature


Figure 4.34: Phase 2.1: History of peak deflections at bottom of steel plate.

42 UCPRC-RR-2011-06

y = 0.0121x - 0.107

R2 = 0.9957

0.0

0.5

1.0

1.5

2.0

2.5

3.0

0 10 20 30 40 50 60 70 80 90 100 110 120 130 140

HVS Wheel Load (kN)

Pea

k V

ert

ica

l Def

lect

ion

(m

m)

Response for loads between 40 and 100kN


Linear Regression

LVDT #5





Peak strain increased with the change in wheel load as expected, and showed similar trends to earlier

phases. Highest strains were recorded on the sensor closest to the wheelpath. After each load change,

peak strain recorded by each of the gauges remained constant, with no evidence of damage accumulation

with increasing load repetitions. The relationship between peak strain and load in this phase was linear for

all three strain gauges for all loads (Figure 4.37 through Figure 4.39 for the three strain gauges). Strain

measurements did not appear to be influenced by temperature in this phase.

4.5.5 Visual Damage



particles (Figure 4.40). Apart from some deformation (approximately 4.0 mm) on the rubber, no damage

was observed on the Trelleborg unit. Permanent deformation is discussed in Section 4.10. No rotation of

the bolts or washers was observed.

4.5.6 Phase Summary


permanent deformation in the steel plate occurred. Responses were similar to those recorded in earlier

phases during loading on the center of the expansion joint. Increases in peak deflection and peak strain

UCPRC-RR-2011-06 43

continued to show a linear relationship with increasing load. Based on the results and observations in this

phase, it was concluded that there was no significant difference in the trends of measurements recorded

during trafficking on the edge compared to those recorded during trafficking on the center. However,

since higher deflections and strains were measured in this phase for the same loads, it was decided to

undertake all further testing on the edge of the bridge deck expansion joint as this was considered more

likely to induce damage.

0

20

40

60

80

100

120

140

160

180

200

0 20 40 60 80 100 120 140 160 180 200


Pea

k S

trai

n (

mic

rost

rain

)

0

5

10

15

20

25

30

35

40

Ste

el P

late

Tem

per

atu

re (

ºC)

SG10SG11SG12Steel plate temperature near SG11


Figure 4.36: Phase 2.1: History of longitudinal strains at bottom of steel plate.

SG #10y = 0.8059x - 5.8753

R2 = 0.9966

0

20

40

60

80

100

120

140

160

180

200

0 10 20 30 40 50 60 70 80 90 100 110 120

HVS Wheel Load (kN)

Pe

ak L

on

git

ud

ina

l S

tra

in (

mic

ros

trai

n)



Linear Regression

Figure 4.37: Phase 2.1: Relationship between peak strains and wheel load for SG #10.


44 UCPRC-RR-2011-06

SG #11

y = 1.2124x - 5.9297

R2 = 0.9964

0

20

40

60

80

100

120

140

160

180

200

0 10 20 30 40 50 60 70 80 90 100 110 120

HVS Wheel Load (kN)

Pea

k L

on

git

ud

inal

Str

ain

(m

icro

str

ain

)



Linear Regression



SG #12

y = 1.8495x - 3.9261

R2 = 0.9979

0

20

40

60

80

100

120

140

160

180

200

0 10 20 30 40 50 60 70 80 90 100 110 120

HVS Wheel Load (kN)

Pe

ak L

on

git

ud

ina

l S

tra

in (

mic

rost

rain

)



Linear Regression



UCPRC-RR-2011-06 45

Figure 4.40: Phase 2.1: Rubber particle accumulation on Trelleborg unit after 928,000 repetitions.

4.6 Phase 3.1: Edge Test with Impact Load and Unidirectional Traffic

4.6.1 Introduction

After reviewing the Phase 1 and Phase 2 results, it was concluded that continued trafficking at 80 kN and

100 kN was unlikely to cause any significant structural damage to the seismic joint in the time available.

The study therefore proceeded to Phase 3 of the test plan, which required impact loading—caused by

including a “step” in the wheelpath—and was followed by significantly heavier wheel loads (using an

aircraft tire).

The objective of Phase 3.1 was to determine whether impact loads on the steel plate had any influence on

the response trends observed during earlier phases. Testing was carried out on the edge of the expansion

joints in the same wheelpath used in Phase 2.1. On the first day, a 13 mm (0.5 in.) neoprene mat

(Figure 4.41) was used to cause the impact and thereafter a 19 mm (0.75 in.) hardwood board

(Figure 4.42). Larger steps could not be used as these would have caused a system error and consequent

shut down of the HVS hydraulic operating unit. The test ran for three days with a 60 kN wheel load in a

unidirectional (i.e., one-way traffic only) channelized mode. Wheel direction travelled from the channel

assembly toward the Trelleborg unit, with impact applied to the channel assembly on the first day

(i.e., from the 13 mm neoprene mat) and then the midpoint of the steel plate thereafter (i.e., from the

19 mm hardwood board. See Figure 4.41 and Figure 4.42 for relative positions of the steps). The

influence of the two steps on actual load applied to the steel plate was not determined.

Rubber accumulation

46 UCPRC-RR-2011-06

Figure 4.41: Phase 3.1: Impact load from neoprene step.

Figure 4.42: Phase 3.1: Impact load from wooden step.

4.6.2 Temperature




Gauge #11 (TC-SG#11). Average ambient temperatures were again typical for the area, had a relatively

small diurnal range, but showed a definite cooling trend compared to the other phases. Average daily

minimum and maximum temperatures recorded on the steel plate were similar to the ambient

temperatures, except for Strain Gauge #12, which again indicated a higher average daily maximum than

the other measurement points, but with a smaller difference compared to the previous phases. No extreme




Thermocouple


Daily Average

Average of Daily

Minimum

Average of Daily

Maximum

Lowest Highest


15 16 15 16 18

13 14 13 14 14

19 18 18 19 22

11 12 12 13 13

22 23 23 24 26


60 60 60 62 64

56 57 56 57 58

66 65 65 66 72

53 54 53 55 56

71 74 74 74 79

UCPRC-RR-2011-06 47

0

5

10

15

20

25

30

35

40

10/3/11 10/4/11 10/5/11 10/6/11 10/7/11 10/8/11

Date

Tem

per

atu

re (

ºC)

0

5

10

15

20

25

30

Lo

ad R

epet

itio

ns

(x 1

,000

)

TC-Ambient

TC-SG#10

TC-SG#11

TC-SG#12

TC-SG#10-S

Number of load repetitions

40kN




are shown in Figure 4.44 through Figure 4.46, respectively. Deflections remained constant for all sensors

throughout the phase, with actual deflection dependent on sensor location in relation to the wheelpath.

Minor fluctuations (~0.01 mm) in deflection for each load were attributed to changes in temperature

and/or load. Based on the data recorded, the impact loads applied did not appear to influence deflection of

the expansion joint at the sensor locations.

0.0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1.0

0 5 10 15 20 25


Pea

k D

efl

ect

ion

(m

m)

0

5

10

15

20

25

30

35

40

Ste

el

Pla

te T

emp

era

ture

(ºC

)

LVDT#1: On Bolt A3LVDT#2: On Bolt B3LVDT#8: On Bolt 4 near TrelleborgSteel plate temperature

60kN


48 UCPRC-RR-2011-06

0.0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1.0

0 5 10 15 20 25


Pea

k D

efl

ecti

on

(m

m)

0

5

10

15

20

25

30

35

40

Ste

el P

late

Te

mp

era

ture

(ºC

)

LVDT#1: On Bolt A3LVDT#2: On Bolt B3LVDT#3: On washer below Bolt A4LVDT#4: On washer below Bolt B4Steel plate temperature

60kN


0.0

0.5

1.0

1.5

2.0

2.5

3.0

3.5

0 5 10 15 20 25


Pea

k D

efl

ecti

on

(m

m)

0

5

10

15

20

25

30

35

40

Ste

el P

late

Tem

pe

ratu

re (

ºC)

LVDT#5: Outside edge at midspanLVDT#6: Inside edge at midspanLVDT#7: Bottom of steel plate at midwidthLVDT#13: Btm. of steel plate, midspan of tunnelSteel plate temperature

60kN



Plots of the peak longitudinal strains measured by the three strain gauges for the duration of this phase are

shown in Figure 4.47. Peak strains remained constant for all sensors throughout the phase, with actual

strain dependent on sensor location in relation to the wheelpath. Minor fluctuations (~2 με) were

attributed to changes in temperature and/or load. Based on the data recorded, the impact loads applied did

not appear to influence strain in the steel plate at the sensor locations.

UCPRC-RR-2011-06 49

0

20

40

60

80

100

120

140

160

180

200

0 5 10 15 20 25


Pea

k S

trai

n (

mic

rost

rain

)

0

5

10

15

20

25

30

35

40

Ste

el P

late

Tem

per

atu

re (

ºC)

SG#10, Outside edgeSG#11, MidwidthSG#12, Inside edgeSteel plate temperature near SG#11

60kN


4.6.5 Visual Damage

No damage was observed to any part of the expansion joint or structure after completion of this phase of

testing. No rotation of the bolts or washers was observed.

4.6.6 Phase Summary

A 60 kN impact load did not appear to influence response in the expansion joint at the sensor locations,

and no damage was observed on completion of this short phase. Responses were similar to those recorded

in earlier phases. There was also no difference observed between unidirectional and bidirectional

trafficking and consequently all further testing was carried out in a bidirectional mode, which applies

more wheel loads than unidirectional trafficking in a given period of time.

4.7 Phase 3.2: Load Response with Impact Load

4.7.1 Introduction

Phase 3.2 assessed load response with impact load by evaluating changes in strain and deflection induced

by increases in wheel load. The test ran for 15 days with five days each at loads of 60 kN, 80 kN, and

100 kN, respectively. All loading was applied to the edge of the expansion joint in a bidirectional

channelized mode. The impact load, which was applied on every alternate pass of the bidirectional

trafficking, was induced with the 19 mm (0.75 in.) hardwood board used in Phase 3.1.

50 UCPRC-RR-2011-06

4.7.2 Temperature

Temperatures were not recorded during this phase due to a data acquisition system malfunction, which

was repaired while testing continued. Given the limited time available to complete the testing and that

temperature appeared to have little or no influence on the behavior of the bridge deck expansion joint, the

project team agreed to continue testing in this phase without temperature data.


Influence lines (or deflection bowls) from a single pass of the 100 kN wheel load (repetition #240,000 for

the phase or #1,210,000 for the test) for the LVDTs on the bolts, washers, and steel plate are shown in

Figure 4.48 through Figure 4.50, respectively. The impact load had a very small effect (wheel position 4.5

in the figures) on response. Plots of the peak deflections measured on bolts, washers, and the steel plate

for the duration of Phase 3.2 are shown in Figure 4.51 through Figure 4.53, respectively. No differences in

behavior to that recorded in Phase 2.1 (edge testing without impact load) were observed, with deflections

remaining constant for each load for all sensors throughout the phase. The relationship between load and

response was linear and consistent with previous phases (Figure 4.54). Based on the data recorded, the

impact applied at any of the wheel loads did not appear to influence deflection of the expansion joint at the

sensor locations.

-0.30

-0.25

-0.20

-0.15

-0.10

-0.05

0.00

0.05

0.10

0.15

0.20

0.25

0.30


Ver

tic

al D

efle

ctio

n (

mm

)

LVDT#1, On Bolt A3 LVDT#2, On Bolt B3 LVDT#8, On Bolt 4 near Trelleborg




(Repetition #1,210,000, wheel load at 100 kN)

UCPRC-RR-2011-06 51

-0.30

-0.25

-0.20

-0.15

-0.10

-0.05

0.00

0.05

0.10

0.15

0.20

0.25

0.30


Ver

tic

al D

efle

cti

on

(m

m)

LVDT#1, On Bolt A3 LVDT#2, On Bolt B3LVDT#3, On washer below Bolt A4 LVDT#4, On washer below Bolt B4





-3.25

-3.00

-2.75

-2.50

-2.25

-2.00

-1.75

-1.50

-1.25

-1.00

-0.75

-0.50

-0.25

0.00

0.25

0.50


Ver

tica

l D

efle

ctio

n (

mm

)







52 UCPRC-RR-2011-06

0.0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1.0

0 25 50 75 100 125 150 175 200 225 250


Pe

ak D

efle

ctio

n (

mm

)LVDT#1: On Bolt A3LVDT#2: On Bolt B3LVDT#8: On Bolt 4 near Trelleborg

60kN 80kN 100kN


0.0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1.0

0 25 50 75 100 125 150 175 200 225 250


Pea

k D

efle

ctio

n (

mm

)

LVDT#1: On Bolt A3LVDT#2: On Bolt B3LVDT#3: On washer below Bolt A4LVDT#4: On Washer below Bolt B4

60kN 80kN 100kN


UCPRC-RR-2011-06 53

0.0

0.5

1.0

1.5

2.0

2.5

3.0

3.5

0 25 50 75 100 125 150 175 200 225 250


Pe

ak D

efl

ecti

on

(m

m)



60kN 80kN 100kN


y = 0.0113x - 0.1287

R2 = 0.9886

0.0

0.5

1.0

1.5

2.0

2.5

3.0

0 10 20 30 40 50 60 70 80 90 100 110 120 130 140

HVS Wheel Load (kN)

Pea

k V

ert

ical

Def

lect

ion

(m

m)

Response for loads between 60 and 100kN

Linear Regression

LVDT #5




Influence lines (or deflection bowls) from a single pass of the 100 kN wheel load (Repetition #240,000 for

the phase and #1,210,000 for the test) for the three strain gauges is shown in Figure 4.55. The impact load

had a very small effect (wheel position 4.5 in the figure) on response. A plot of the peak strains measured

54 UCPRC-RR-2011-06

on bolts, washers, and the steel plate for the duration of Phase 3.2 are shown in Figure 4.56. No

differences in behavior to that recorded in Phase 2.1 were observed, with strains remaining constant for

each load for all sensors throughout the phase. The relationship between load and response was linear and

consistent with previous phases for all three sensors (Figure 4.57 through Figure 4.59). Based on the data

recorded, the impact applied at any of the wheel loads did not appear to influence longitudinal strain on

the steel plate at the sensor locations.

-25

0

25

50

75

100

125

150

175

200

0 1 2 3 4 5 6 7 8

Wheel Position (m)

Lo

ng

itu

din

al

Str

ain

(m

icro

stra

in)

SG#10, Outside edgeSG#11, MidwidthSG#12, Inside edge

Positive strain = Tension

Trelleborg Steel Deck ConcreteConcrete

Figure 4.55: Phase 3.2: Influence lines for longitudinal strains at bottom of steel plate.


0

20

40

60

80

100

120

140

160

180

200

0 25 50 75 100 125 150 175 200 225 250


Pea

k S

tra

in (

mic

rost

rain

)

SG#10, Outside edge SG#11, Midwidth SG#12, Inside edge

60kN 80kN 100kN


UCPRC-RR-2011-06 55

SG #10y = 0.7699x - 6.0358

R2 = 0.9891

0

20

40

60

80

100

120

140

160

180

200

0 10 20 30 40 50 60 70 80 90 100 110 120

HVS Wheel Load (kN)

Pea

k L

on

git

ud

inal

Str

ain

(m

icro

stra

in) Response for loads between 60kN and 100kN

Linear Regression



SG #11y = 1.1716x - 7.0123

R2 = 0.9889

0

20

40

60

80

100

120

140

160

180

200

0 10 20 30 40 50 60 70 80 90 100 110 120

HVS Wheel Load (kN)

Pea

k L

on

git

ud

inal

Str

ain

(m

icro

stra

in) Response for loads between 60kN and 100kN

Linear Regression



56 UCPRC-RR-2011-06

SG #12

y = 1.7916x - 5.3185

R2 = 0.9893

0

20

40

60

80

100

120

140

160

180

200

0 10 20 30 40 50 60 70 80 90 100 110 120

HVS Wheel Load (kN)

Pea

k L

on

git

ud

ina

l S

trai

n (

mic

rost

rain

)


Linear Regression



4.7.5 Visual Damage



particles (Figure 4.60). Apart from some additional deformation on the rubber, no new damage was

observed on the Trelleborg unit. No rotation of the bolts or washers was observed.

Figure 4.60: Phase 3.2: Rubber accumulation on Trelleborg unit after 1,191,000 repetitions.

4.7.6 Phase Summary


permanent deformation in the steel plate occurred. Responses continued to be the same as those recorded

Start of rubber accumulation

UCPRC-RR-2011-06 57

in earlier phases and increases in peak deflection and peak strain continued to show a linear relationship

with increasing load. Based on the results and observations in this phase, it was concluded that the impact

loading applied did not significantly influence performance.

4.8 Phase 3.3: Edge Test with High Load

4.8.1 Introduction

The objective of Phase 3.3 was to cause as much damage to the joint as possible to identify the weakest

part of the design. The test ran for 15 days using an aircraft tire, with one day of half-axle loading at

100 kN and 14 days at 150 kN. All loading was applied to the edge of the expansion joint in a

bidirectional channelized mode.

4.8.2 Temperature




Gauge #11 (TC-SG#11). Average ambient temperatures were again typical for the area, had a relatively

small diurnal range, and continued to show the cooling trend observed in Phase 3.1, but with a number of

unseasonably warm days. Average daily minimum and maximum temperatures recorded on the steel plate

were similar to the ambient temperatures, except for the thermocouple on Strain Gauge #12, which again

indicated a higher average daily maximum than the other measurement points. No extreme temperature

events were recorded. It is unlikely that temperature had any significant influence on the way that the

bridge deck expansion joint components functioned during this phase of testing.


Thermocouple


Daily Average

Average of Daily

Minimum

Average of Daily

Maximum

Lowest Highest


18 19 18 21 26

14 16 15 17 20

22 21 21 25 33

11 13 12 15 18

28 23 23 27 37


64 65 65 70 78

58 61 59 63 68

72 70 69 76 91

52 55 54 58 64

83 74 73 81 98

58 UCPRC-RR-2011-06

0

5

10

15

20

25

30

35

40

10/18/11 10/20/11 10/22/11 10/24/11 10/26/11 10/28/11 10/30/11 11/1/11 11/3/11 11/5/11

Date

Tem

per

atu

re (

ºC)

0

25

50

75

100

125

150

175

200

Lo

ad

Rep

etit

ion

s (

x 1

,00

0)

TC-Ambient

TC-SG#10

TC-SG#11

TC-SG#12

TC-SG#10-S

Number of load repetitions

150kN100kN




are shown in Figure 4.62 through Figure 4.64, respectively. Deflections remained constant for all sensors

throughout the phase, with actual deflection dependent on sensor location in relation to the wheelpath.

Minor fluctuations (~0.01 mm) in deflection were attributed to changes in temperature and/or load and

were consistent with previous test phases. Based on the data recorded, the very high loads applied did not

appear to influence deflection of the expansion joint at the sensor locations and there was no evidence of

any accumulated damage.

0.0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1.0

0 20 40 60 80 100 120 140 160Accumulated Load Repetitions (x 1,000)

Pea

k D

efl

ec

tio

n (

mm

)

0

5

10

15

20

25

30

35

40

Ste

el P

late

Te

mp

erat

ure

(ºC

)

LVDT#1: On Bolt A3 LVDT#2: On Bolt B3

LVDT#8: On Bolt 4 near Trelleborg Steel plate temperature

100kN 150kN


UCPRC-RR-2011-06 59

0.0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1.0

0 20 40 60 80 100 120 140 160


Pe

ak D

efle

cti

on

(m

m)

0

5

10

15

20

25

30

35

40

Ste

el

Pla

te T

em

per

atu

re (

ºC)


100kN 150kN


0.0

0.5

1.0

1.5

2.0

2.5

3.0

3.5

4.0

4.5

5.0

0 20 40 60 80 100 120 140 160


Pea

k D

efle

ctio

n (

mm

)

0

5

10

15

20

25

30

35

40

Ste

el P

late

Tem

per

atu

re (

ºC)



Steel plate temperature

100kN 150kN



A plot of the peak strains measured on the steel plate for the duration of Phase 3.3 is shown in Figure 4.65.

Strains remained constant for all sensors throughout the phase and were consistent with those measured in

previous phases. Increasing the load to 150 kN resulted in a linear increase in strain recorded at the

various sensors. Minor fluctuations (<5 με) in longitudinal strain were again attributed to changes in

temperature and/or load and were consistent with previous phases. Based on the data recorded, the very

60 UCPRC-RR-2011-06

high loads applied did not appear to influence longitudinal strain in the steel plate at the sensor locations

and there was no evidence of any accumulated damage.

0

25

50

75

100

125

150

175

200

225

250

275

300

0 20 40 60 80 100 120 140 160


Pea

k S

tra

in (

mic

ros

trai

n)

0

5

10

15

20

25

30

35

40

Ste

el P

late

Tem

per

atu

re (

ºC)

SG#10, Outside edge SG#11, Midwidth SG#12, Inside edge Steel plate temperature

100kN 150kN


4.8.5 Visual Damage

No visual damage was observed on the steel plate, bolts, or washers at the end of this phase. However,

isolated damage was caused by the very high wheel load to the steel ribs of the Trelleborg unit directly

under the wheelpath (Figure 4.66). Damage to these ribs then resulted in severe deformation/shoving

(Figure 4.67) and then tearing of the rubber (Figure 4.68). Large quantities of accumulated rubber

particles were observed between the ribs and in other depressions. The concrete approach slab also

cracked under the very heavy loading, but this did not influence the behavior of the expansion joint in any

way. No rotation of the bolts or washers was observed. Photographs of the structure and Trelleborg unit

on completion of all testing are provided in Figure 4.69.

4.8.6 Phase Summary

On completion of this phase of testing, the measured data from the LVDTs and strain gauges indicated

that there was still no structural damage on any steel parts of the expansion joint. Observed damage was

limited to the wheelpath over the Trelleborg unit only, and consisted of significant wear and deformation

on the rubber sections of the Trelleborg unit and deformation and shearing in one of the steel ribs

supporting these rubber sections. The distress observed to the Trelleborg unit under the very high loads

(almost four times the legal limit) applied in this last phase of testing is unlikely to occur under normal

traffic on the Bay Bridge.

UCPRC-RR-2011-06 61

Figure 4.66: Phase 3.3: Damage to steel rib of Trelleborg unit.

Figure 4.67: Phase 3.3: Deformation and shoving of rubber on Trelleborg unit.

Figure 4.68: Phase 3.3: Tearing of rubber and accumulation of rubber particles in Trelleborg unit.

Figure 4.69: Phase 3.3: Structure and Trelleborg unit after completion of testing.

Rubber accumulation

62 UCPRC-RR-2011-06

4.9 Static Responses for All Phases

Static responses are those deflections and strains measured on the bolts, washers, and steel plate caused by

either temperature change or plastic deformation induced by wheel loading. They are termed “static”

because their rate of change is much slower compared to the dynamic responses of the moving wheels

discussed in Section 4.2 through Section 4.8 above.

4.9.1 Vertical Deflections

Example daily variation in vertical deflections measured during Phase 1.1 (representative of all phases) is

shown in Figure 4.70. Deflection increased with increasing temperature and generally followed daily

temperature change. However, the amount of change in any day was miniscule (~0.03 mm) and was not

considered significant to the study.

-0.15

-0.10

-0.05

0.00

0.05

0.10

0.15

8/10/11 8/11/11 8/12/11 8/13/11 8/14/11 8/15/11 8/16/11 8/17/11

Date

Sta

tic

Ver

tic

al D

efle

cti

on

(m

m)

0

5

10

15

20

25

30

35

40

Ste

el

Pla

te T

em

pe

ratu

re (

ºC)

LVDT#5 LVDT#8 LVDT#7 TC-SG#11

Figure 4.70: Example daily variation for vertical deflections during Phase 1.1.

The history of daily maximum static vertical deflections measured on the steel plate and daily maximum

steel plate temperature is shown in Figure 4.71. Deflections did not show strong correlation with

temperature. However, some very small change in daily maximum static vertical deflection is evident over

the duration of the study (between 0.1 mm and 0.3 mm). It is not clear whether this could be attributed to

settlement of the structure on the soft clay subgrade or to permanent deformation caused by the very heavy

wheel loads. The amount of change in vertical deflection did not warrant further investigation and was

not considered significant to the study.

UCPRC-RR-2011-06 63

-0.5

-0.4

-0.3

-0.2

-0.1

0

0.1

0.2

0.3

0.4

0.5

8/8/2011 8/22/2011 9/5/2011 9/19/2011 10/3/2011 10/17/2011 10/31/2011

Date

Da

ily M

axim

um

Sta

tic

Def

lect

ion

(m

m)

0

5

10

15

20

25

30

35

40

Dai

ly M

axim

um

Ste

el P

late

Tem

per

atu

re (

ºC)

LVDT#5, Outside edge LVDT#6, Inside edgeLVDT#7, Midwidth LVDT#10, Midspan of tunnelTC-SG#11

Phases 1.1 to 1.3 Phase 2.1 Phases 3.1 & 3.2 Phase 3.3

Decreasing deflection is downward movement / compression

Figure 4.71: History of daily maximum static vertical deflections on steel plate.


Example daily variations in longitudinal strain for the three strain gauges, measured during Phase 1.1

(representative of all phases), are shown in Figure 4.72 through Figure 4.74. Strain increased with

increasing temperature and generally followed (with a lag) daily temperature change. However, the

amount of change in any day was very small (~30 με to 70 με) and was not considered significant to the

study.

SG #10, Outside edge

-60

-50

-40

-30

-20

-10

0

10

20

30

40

8/10/11 8/11/11 8/12/11 8/13/11 8/14/11 8/15/11 8/16/11 8/17/11

Date

Sta

tic

Lo

ng

itu

din

al S

trai

n

(mic

ros

trai

n)

0

5

10

15

20

25

30

35

40

Ste

el

Pla

te T

em

per

atu

re (

ºC)

SG#10 TC-SG#10

Increasing strain is tensile strain

Figure 4.72: Example daily variation for longitudinal strain at SG #10 during Phase 1.1.

64 UCPRC-RR-2011-06

SG #11, Midspan

-60

-50

-40

-30

-20

-10

0

10

20

30

40

8/10/11 8/11/11 8/12/11 8/13/11 8/14/11 8/15/11 8/16/11 8/17/11

Date

Sta

tic

Lo

ng

itu

din

al S

trai

n

(mic

rost

rain

)

0

5

10

15

20

25

30

35

40

Ste

el P

late

Tem

per

atu

re (

ºC)

SG#11 TC-SG#11



SG # 12, Inside edge

-60

-50

-40

-30

-20

-10

0

10

20

30

40

8/10/11 8/11/11 8/12/11 8/13/11 8/14/11 8/15/11 8/16/11 8/17/11

Date

Sta

tic

Lo

ng

itu

din

al S

trai

n

(mic

rost

rain

)

0

5

10

15

20

25

30

35

40

Ste

el P

late

Tem

per

atu

re (

ºC)

SG#12 TC-SG#12



Figure 4.75 shows the history of daily maximum static longitudinal strains as well as daily maximum steel

plate temperatures for all phases of HVS testing. Changes in daily maximum longitudinal strain generally

followed changes in temperature, and any permanent strain caused by the wheel loading is considered to

be insignificant.

UCPRC-RR-2011-06 65

-40

-30

-20

-10

0

10

20

30

40

8/8/2011 8/22/2011 9/5/2011 9/19/2011 10/3/2011 10/17/2011 10/31/2011

Date

Dai

ly M

axi

mu

m S

tati

c S

trai

n (

mic

rost

rain

)

0

5

10

15

20

25

30

35

40

Dai

ly M

axim

um

Ste

el P

late

Tem

per

atu

re (

ºC)

SG#10, Outside edge

SG#11, Midwidth

SG#12, Inside edge

TC-SG#11

Phases 1.1 to 1.3 Phase 2.1 Phases 3.1 & 3.2 Phase 3.3

Decreasing strain is compression

Figure 4.75: History of daily maximum static longitudinal strains.

4.10 Permanent Deformation on Trelleborg Unit

The history of maximum downward permanent deformation on the Trelleborg unit at different

profilometer measurement stations is shown in Figure 4.76. Scatter on the figure is attributed to the

irregular surface of the Trelleborg unit, the resilient properties of the material from which it is constructed,

and constant changes to the shape of the rubber caused by tire load and accumulating damage. Permanent

deformations of about 1.0 mm were recorded at Station 8 even though this station was approximately

300 mm (12 in.) from the edge of the wheelpath, where no actual deformation should have been recorded.

This implies that the accuracy of the laser profilometer measurements on the Trelleborg unit is about

1.0 mm (0.04 in.).

Figure 4.77 shows the history of average maximum downward permanent deformation of the Trelleborg

unit for all stations (Station 1 through Station 8). The permanent deformation increased to approximately

2.5 mm (0.1 in.) after only 20,000 repetitions (which were applied with a 25 kN load), then recovered to

approximately 1.0 mm (0.04 in.) after 60,000 repetitions with a 40 kN load, and then remained relatively

constant until it started steadily increasing again at the end of Phase 1.3. The reason for the initial increase

in permanent deformation during Phase 1.1 is unclear but is considered to be insignificant since it

recovered under further trafficking.

66 UCPRC-RR-2011-06

0

1

2

3

4

5

6

7

8

9

10

0 200 400 600 800 1,000 1,200 1,400


Max

imu

m D

ow

nw

ard

Def

orm

ati

on

(m

m)

Station 1 Station 2 Station 3 Station 4 Station 5 Station 6 Station 7 Station 8

Ph

ases

1.1

an

d 1

.2

Pha

se 1

.3

Pha

se 2

.1

Pha

se 3

.1

Pha

se 3

.2

Pha

se 3

.3

Figure 4.76: Maximum downward permanent deformation of the Trelleborg unit.

0.0

0.5

1.0

1.5

2.0

2.5

3.0

3.5

4.0

2000

037

004

5650

665

008

8200

010

3007

1180

3213

7002

1565

2817

6696

1930

0021

6102

2330

0025

0006

2700

0729

0136

3102

0032

8015

3481

0036

7000

3850

0040

5000

4240

0044

5000

4630

0848

3000

5010

0051

8014

5380

1655

6500

5780

0059

8000

6180

0263

8000

1910

037

800

6000

274

019

9520

110

3000

1210

0013

601

2800

036

000

5612

876

000

9470

811

6000

1360

0015

3000

1686

3118

8843

7888

2303

610

0340

000

5600

097

079

1139

0912

8915

1425

0016

0000

1775

0019

3009

2065

0822

2114

2400

0820

008

3850

857

000

7501

093

000

1210

0013

2925

1500

08

1.1 and 1.2 1.3 2.1 3.1 3.2 3.3

Phase and Repetition

Av

erag

e D

ow

nw

ard

Def

orm

atio

n (

mm

)

Figure 4.77: Average maximum downward permanent deformation of Trelleborg unit.

UCPRC-RR-2011-06 67

Severe damage to the Trelleborg unit was first observed after about 1,230,000 load repetitions, or after

about 20,000 load repetitions into Phase 3.3 testing with the aircraft tire (Figure 4.66 through Figure 4.68).

Vertical permanent deformation contour plots at the end of each phase are shown in Figure 4.78 through

Figure 4.85. These contour plots show how the surface elevation changed at different profilometer

measurement locations. Note that the permanent deformation between profilometer measurement stations

is linearly interpolated and that scales are different on each plot.

Stations

Tra

nsve

rse

Dis

tanc

e (m

m)

Phase 1.1, Repetition = 0.518 Million (Final)

0 2 4 6 8

-500

0

500

1000

1500

Color Map for Profilometer Reading (mm)-2.6 -1.9 -1.2 -0.5 0.2 0.9 1.6

Figure 4.78: Phase 1.1: Contour plot of deformation (dual wheel, channelized on center).

(Max. downward deformation of 2.6 mm [Station 5], max. upward deformation of 1.6 mm [Station 2])

Stations

Tra

nsve

rse

Dis

tanc

e (m

m)


0 2 4 6 8

-500

0

500

1000

1500


Figure 4.79: Phase 1.2: Contour plot of deformation (dual wheel, channelized on center).


Trelleborg

Trelleborg

Wheelpath

Wheelpath

68 UCPRC-RR-2011-06

Stations

Tra

nsve

rse

Dis

tanc

e (m

m)


0 2 4 6 8

-500

0

500

1000

1500

Color Map for Profilometer Reading (mm)-2.0 -1.2 -0.4 0.3 1.1 1.9 2.6

Figure 4.80: Phase 1.3: Contour plot of deformation (dual wheel, wander).


Stations

Tra

nsve

rse

Dis

tanc

e (m

m)


0 2 4 6 8

-500

0

500

1000

1500

Color Map for Profilometer Reading (mm)-4.8 -3.6 -2.5 -1.4 -0.3 0.8 2.0

Figure 4.81: Phase 2.1: Contour plot of deformation (dual wheel, channelized on edge).


Trelleborg

Trelleborg

Wheelpath

Wheelpath

UCPRC-RR-2011-06 69

Stations

Tra

nsve

rse

Dis

tanc

e (m

m)

Phase 3.1, Repetition = 23.036 Thousand (Final)

0 2 4 6 8

-500

0

500

1000

1500


Figure 4.82: Phase 3.1: Contour plot of deformation (dual wheel, channelized on edge).


Stations

Tra

nsve

rse

Dis

tanc

e (m

m)


0 2 4 6 8

-500

0

500

1000

1500


Figure 4.83: Phase 3.2: Contour plot of deformation (dual wheel, channelized on edge with impact).


Trelleborg

Trelleborg

Wheelpath

Wheelpath

70 UCPRC-RR-2011-06

Stations

Tra

nsve

rse

Dis

tanc

e (m

m)

Phase 3.3, Repetition = 20.008 Thousand

0 2 4 6 8

-500

0

500

1000

1500


Figure 4.84: Phase 3.3 (20,000 reps): Contour plot of deformation (aircraft, channelized on edge).


Stations

Tra

nsve

rse

Dis

tanc

e (m

m)


0 2 4 6 8

-500

0

500

1000

1500


Figure 4.85: Phase 3.3 (final): Contour plot of deformation (aircraft, channelized on edge).


Trelleborg

Trelleborg

Wheelpath

Wheelpath

UCPRC-RR-2011-06 71

5. CONCLUSIONS

A relatively unique opportunity was recently identified for accelerated traffic load testing of a new bridge

expansion joint design not previously used in California. This study was part of the construction of the

new East Span of the San Francisco–Oakland Bay Bridge and assessed whether the new expansion joints

(which were designed to function in harmony with the bridge decks in the event of a high-magnitude

earthquake) planned for linking the Self-anchored Span with the Transition and Skyway spans would

withstand truck traffic loading. A test structure incorporating one of the full-scale joints was constructed

close to the actual bridge and tested with the California Department of Transportation / University of

California Pavement Research Center Heavy Vehicle Simulator in a series of phases.

A total of 1.36 million load repetitions, equating to about 46 million equivalent standard axle loads on a

highway pavement, were applied in seven phases during the three-month test. On completion of this

testing, no structural damage was recorded by any of the Linear Variable Differential Transducers

(LVDTs) or strain gauges installed on the steel plates, steel frames, bolts, and washers. There was also no

visible damage on any of these components. Excessive overloading with a 150 kN half-axle load

(approximately four times the standard axle load) on an aircraft tire in the last phase of the test caused

some damage to the Trelleborg unit in the joint. The damage included abrasion, tearing, shoving and

permanent deformation of the rubber inserts, and deformation and shearing of one of the steel supports

directly under the wheel load.

Although no vehicle suspension dynamics (i.e., vehicle bounce) or speed effects were considered, based

on the results of this limited testing, it was concluded that the Caltrans seismic expansion joint would

perform adequately under typical Bay Bridge traffic. The distresses observed on the Trelleborg unit under

high loads in the last phase of testing are unlikely to occur under normal traffic. However, the Trelleborg

unit was found to be the weakest point of the expansion joint, as expected. On the actual bridge structure,

these units should be checked periodically to confirm the findings of this study, and to assess any effects

of higher speeds and vehicle dynamics that were not identified. The joints will require periodic

maintenance and replacement in line with manufacturer’s specifications.

The findings from this study indicate that the Caltrans seismic expansion joint tested would be

appropriate for typical Bay Bridge traffic.

No seismic or structural testing was undertaken and no recommendations toward the expansion joint’s

seismic or structural performance are made. Ride quality, skid resistance, and tire noise studies were

carried out by Caltrans in a separate study and are reported on in separate Caltrans reports.

72 UCPRC-RR-2011-06

UCPRC-RR-2011-06 73

6. REFERENCES

1. JONES, D. 2005. Quality Management System for Site Establishment, Daily Operations,

Instrumentation, Data Collection and Data Storage for APT Experiments. Pretoria, South

Africa: CSIR Transportek. (Contract Report CR-2004/67-v2).

74 UCPRC-RR-2011-06

UCPRC-RR-2011-06 75

APPENDIX A: TEST STRUCTURE DESIGN AND INSTRUMENTATION

76 UCPRC-RR-2011-06

UCPRC-RR-2011-06 77

Accelerated Traffic Load Testing of Expansion Joints for ... · perform adequately under typical Bay Bridge traffic. The distresses observed on the Trelleborg unit under high loads

Documents